117 files changed, 831 insertions, 260 deletions
diff --git a/src/common/CMakeLists.txt b/src/common/CMakeLists.txt
index 6ffc612e7..d1ec8ff08 100644
--- a/src/common/CMakeLists.txt
+++ b/src/common/CMakeLists.txt
@@ -172,5 +172,5 @@ endif()
 create_target_directory_groups(common)
-target_link_libraries(common PUBLIC Boost::boost fmt microprofile)
+target_link_libraries(common PUBLIC Boost::boost fmt::fmt microprofile)
-target_link_libraries(common PRIVATE lz4_static libzstd_static)
+target_link_libraries(common PRIVATE lz4::lz4 zstd::zstd)
diff --git a/src/common/uuid.h b/src/common/uuid.h
index f6ad064fb..4d3af8cec 100644
--- a/src/common/uuid.h
+++ b/src/common/uuid.h
@@ -40,6 +40,11 @@ struct UUID {
        uuid = INVALID_UUID;
    }
+    // TODO(ogniK): Properly generate a Nintendo ID
+    constexpr u64 GetNintendoID() const {
+        return uuid[0];
+    }
    std::string Format() const;
    std::string FormatSwitch() const;
 };
diff --git a/src/core/CMakeLists.txt b/src/core/CMakeLists.txt
index 8546d3602..47418006b 100644
--- a/src/core/CMakeLists.txt
+++ b/src/core/CMakeLists.txt
@@ -606,11 +606,11 @@ endif()
 create_target_directory_groups(core)
 target_link_libraries(core PUBLIC common PRIVATE audio_core video_core)
-target_link_libraries(core PUBLIC Boost::boost PRIVATE fmt json-headers mbedtls opus unicorn)
+target_link_libraries(core PUBLIC Boost::boost PRIVATE fmt::fmt nlohmann_json::nlohmann_json mbedtls Opus::Opus unicorn)
 if (YUZU_ENABLE_BOXCAT)
    target_compile_definitions(core PRIVATE -DYUZU_ENABLE_BOXCAT)
-    target_link_libraries(core PRIVATE httplib json-headers zip)
+    target_link_libraries(core PRIVATE httplib nlohmann_json::nlohmann_json zip)
 endif()
 if (ENABLE_WEB_SERVICE)
diff --git a/src/core/hle/kernel/handle_table.cpp b/src/core/hle/kernel/handle_table.cpp
index e441a27fc..35448b576 100644
--- a/src/core/hle/kernel/handle_table.cpp
+++ b/src/core/hle/kernel/handle_table.cpp
@@ -30,6 +30,7 @@ HandleTable::~HandleTable() = default;
 ResultCode HandleTable::SetSize(s32 handle_table_size) {
    if (static_cast<u32>(handle_table_size) > MAX_COUNT) {
+        LOG_ERROR(Kernel, "Handle table size {} is greater than {}", handle_table_size, MAX_COUNT);
        return ERR_OUT_OF_MEMORY;
    }
@@ -80,6 +81,7 @@ ResultVal<Handle> HandleTable::Duplicate(Handle handle) {
 ResultCode HandleTable::Close(Handle handle) {
    if (!IsValid(handle)) {
+        LOG_ERROR(Kernel, "Handle is not valid! handle={:08X}", handle);
        return ERR_INVALID_HANDLE;
    }
diff --git a/src/core/hle/kernel/hle_ipc.cpp b/src/core/hle/kernel/hle_ipc.cpp
index 91d94025c..ba0eac4c2 100644
--- a/src/core/hle/kernel/hle_ipc.cpp
+++ b/src/core/hle/kernel/hle_ipc.cpp
@@ -13,7 +13,6 @@
 #include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "common/logging/log.h"
-#include "core/core.h"
 #include "core/hle/ipc_helpers.h"
 #include "core/hle/kernel/handle_table.h"
 #include "core/hle/kernel/hle_ipc.h"
@@ -57,7 +56,6 @@ std::shared_ptr<WritableEvent> HLERequestContext::SleepClientThread(
            return true;
        });
-    auto& kernel = Core::System::GetInstance().Kernel();
    if (!writable_event) {
        // Create event if not provided
        const auto pair = WritableEvent::CreateEventPair(kernel, "HLE Pause Event: " + reason);
@@ -79,9 +77,11 @@ std::shared_ptr<WritableEvent> HLERequestContext::SleepClientThread(
    return writable_event;
 }
-HLERequestContext::HLERequestContext(std::shared_ptr<Kernel::ServerSession> server_session,
+HLERequestContext::HLERequestContext(KernelCore& kernel, Core::Memory::Memory& memory,
+                                     std::shared_ptr<ServerSession> server_session,
                                     std::shared_ptr<Thread> thread)
-    : server_session(std::move(server_session)), thread(std::move(thread)) {
+    : server_session(std::move(server_session)),
+      thread(std::move(thread)), kernel{kernel}, memory{memory} {
    cmd_buf[0] = 0;
 }
@@ -216,7 +216,6 @@ ResultCode HLERequestContext::PopulateFromIncomingCommandBuffer(const HandleTabl
 ResultCode HLERequestContext::WriteToOutgoingCommandBuffer(Thread& thread) {
    auto& owner_process = *thread.GetOwnerProcess();
    auto& handle_table = owner_process.GetHandleTable();
-    auto& memory = Core::System::GetInstance().Memory();
    std::array<u32, IPC::COMMAND_BUFFER_LENGTH> dst_cmdbuf;
    memory.ReadBlock(owner_process, thread.GetTLSAddress(), dst_cmdbuf.data(),
@@ -286,7 +285,6 @@ std::vector<u8> HLERequestContext::ReadBuffer(std::size_t buffer_index) const {
    std::vector<u8> buffer;
    const bool is_buffer_a{BufferDescriptorA().size() > buffer_index &&
                           BufferDescriptorA()[buffer_index].Size()};
-    auto& memory = Core::System::GetInstance().Memory();
    if (is_buffer_a) {
        ASSERT_MSG(BufferDescriptorA().size() > buffer_index,
@@ -319,7 +317,6 @@ std::size_t HLERequestContext::WriteBuffer(const void* buffer, std::size_t size,
        size = buffer_size; // TODO(bunnei): This needs to be HW tested
    }
-    auto& memory = Core::System::GetInstance().Memory();
    if (is_buffer_b) {
        ASSERT_MSG(BufferDescriptorB().size() > buffer_index,
                   "BufferDescriptorB invalid buffer_index {}", buffer_index);
diff --git a/src/core/hle/kernel/hle_ipc.h b/src/core/hle/kernel/hle_ipc.h
index af3330297..b31673928 100644
--- a/src/core/hle/kernel/hle_ipc.h
+++ b/src/core/hle/kernel/hle_ipc.h
@@ -19,6 +19,10 @@
 union ResultCode;
+namespace Core::Memory {
+class Memory;
+}
 namespace Service {
 class ServiceFrameworkBase;
 }
@@ -28,6 +32,7 @@ namespace Kernel {
 class Domain;
 class HandleTable;
 class HLERequestContext;
+class KernelCore;
 class Process;
 class ServerSession;
 class Thread;
@@ -98,7 +103,8 @@ protected:
 */
 class HLERequestContext {
 public:
-    explicit HLERequestContext(std::shared_ptr<ServerSession> session,
+    explicit HLERequestContext(KernelCore& kernel, Core::Memory::Memory& memory,
+                               std::shared_ptr<ServerSession> session,
                               std::shared_ptr<Thread> thread);
    ~HLERequestContext();
@@ -305,6 +311,9 @@ private:
    std::vector<std::shared_ptr<SessionRequestHandler>> domain_request_handlers;
    bool is_thread_waiting{};
+    KernelCore& kernel;
+    Core::Memory::Memory& memory;
 };
 } // namespace Kernel
diff --git a/src/core/hle/kernel/memory/address_space_info.cpp b/src/core/hle/kernel/memory/address_space_info.cpp
index 27fae05e7..a523a2502 100644
--- a/src/core/hle/kernel/memory/address_space_info.cpp
+++ b/src/core/hle/kernel/memory/address_space_info.cpp
@@ -2,8 +2,8 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-// This file references various implementation details from Atmosph�re, an open-source firmware for
+// This file references various implementation details from Atmosphere, an open-source firmware for
-// the Nintendo Switch. Copyright 2018-2020 Atmosph�re-NX.
+// the Nintendo Switch. Copyright 2018-2020 Atmosphere-NX.
 #include <array>
@@ -49,18 +49,18 @@ constexpr bool IsAllowedIndexForAddress(std::size_t index) {
    return index < std::size(AddressSpaceInfos) && AddressSpaceInfos[index].GetAddress() != Invalid;
 }
-constexpr std::size_t
+constexpr std::array<std::size_t, static_cast<std::size_t>(AddressSpaceInfo::Type::Count)>
-    AddressSpaceIndices32Bit[static_cast<std::size_t>(AddressSpaceInfo::Type::Count)]{
+    AddressSpaceIndices32Bit{
        0, 1, 0, 2, 0, 3,
    };
-constexpr std::size_t
+constexpr std::array<std::size_t, static_cast<std::size_t>(AddressSpaceInfo::Type::Count)>
-    AddressSpaceIndices36Bit[static_cast<std::size_t>(AddressSpaceInfo::Type::Count)]{
+    AddressSpaceIndices36Bit{
        4, 5, 4, 6, 4, 7,
    };
-constexpr std::size_t
+constexpr std::array<std::size_t, static_cast<std::size_t>(AddressSpaceInfo::Type::Count)>
-    AddressSpaceIndices39Bit[static_cast<std::size_t>(AddressSpaceInfo::Type::Count)]{
+    AddressSpaceIndices39Bit{
        9, 8, 8, 10, 12, 11,
    };
diff --git a/src/core/hle/kernel/memory/address_space_info.h b/src/core/hle/kernel/memory/address_space_info.h
index cc9a6421e..c479890be 100644
--- a/src/core/hle/kernel/memory/address_space_info.h
+++ b/src/core/hle/kernel/memory/address_space_info.h
@@ -2,12 +2,11 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-// This file references various implementation details from Atmosph�re, an open-source firmware for
+// This file references various implementation details from Atmosphere, an open-source firmware for
-// the Nintendo Switch. Copyright 2018-2020 Atmosph�re-NX.
+// the Nintendo Switch. Copyright 2018-2020 Atmosphere-NX.
 #pragma once
-#include "common/common_funcs.h"
 #include "common/common_types.h"
 namespace Kernel::Memory {
diff --git a/src/core/hle/kernel/memory/memory_block.h b/src/core/hle/kernel/memory/memory_block.h
index 9db1f7b39..9d7839d08 100644
--- a/src/core/hle/kernel/memory/memory_block.h
+++ b/src/core/hle/kernel/memory/memory_block.h
@@ -2,8 +2,8 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-// This file references various implementation details from Atmosph�re, an open-source firmware for
+// This file references various implementation details from Atmosphere, an open-source firmware for
-// the Nintendo Switch. Copyright 2018-2020 Atmosph�re-NX.
+// the Nintendo Switch. Copyright 2018-2020 Atmosphere-NX.
 #pragma once
diff --git a/src/core/hle/kernel/memory/memory_block_manager.cpp b/src/core/hle/kernel/memory/memory_block_manager.cpp
index 900395c37..0732fa5a1 100644
--- a/src/core/hle/kernel/memory/memory_block_manager.cpp
+++ b/src/core/hle/kernel/memory/memory_block_manager.cpp
@@ -67,7 +67,6 @@ void MemoryBlockManager::Update(VAddr addr, std::size_t num_pages, MemoryState p
                                MemoryPermission prev_perm, MemoryAttribute prev_attribute,
                                MemoryState state, MemoryPermission perm,
                                MemoryAttribute attribute) {
-    const std::size_t prev_count{memory_block_tree.size()};
    const VAddr end_addr{addr + num_pages * PageSize};
    iterator node{memory_block_tree.begin()};
@@ -109,7 +108,6 @@ void MemoryBlockManager::Update(VAddr addr, std::size_t num_pages, MemoryState p
 void MemoryBlockManager::Update(VAddr addr, std::size_t num_pages, MemoryState state,
                                MemoryPermission perm, MemoryAttribute attribute) {
-    const std::size_t prev_count{memory_block_tree.size()};
    const VAddr end_addr{addr + num_pages * PageSize};
    iterator node{memory_block_tree.begin()};
@@ -145,7 +143,6 @@ void MemoryBlockManager::Update(VAddr addr, std::size_t num_pages, MemoryState s
 void MemoryBlockManager::UpdateLock(VAddr addr, std::size_t num_pages, LockFunc&& lock_func,
                                    MemoryPermission perm) {
-    const std::size_t prev_count{memory_block_tree.size()};
    const VAddr end_addr{addr + num_pages * PageSize};
    iterator node{memory_block_tree.begin()};
diff --git a/src/core/hle/kernel/memory/memory_block_manager.h b/src/core/hle/kernel/memory/memory_block_manager.h
index 9451b5df6..6e1d41075 100644
--- a/src/core/hle/kernel/memory/memory_block_manager.h
+++ b/src/core/hle/kernel/memory/memory_block_manager.h
@@ -6,7 +6,6 @@
 #include <functional>
 #include <list>
-#include <memory>
 #include "common/common_types.h"
 #include "core/hle/kernel/memory/memory_block.h"
diff --git a/src/core/hle/kernel/memory/memory_manager.cpp b/src/core/hle/kernel/memory/memory_manager.cpp
index 3cd4f9e85..6b432e1b2 100644
--- a/src/core/hle/kernel/memory/memory_manager.cpp
+++ b/src/core/hle/kernel/memory/memory_manager.cpp
@@ -104,9 +104,9 @@ ResultCode MemoryManager::Allocate(PageLinkedList& page_list, std::size_t num_pa
    // Ensure that we don't leave anything un-freed
    auto group_guard = detail::ScopeExit([&] {
        for (const auto& it : page_list.Nodes()) {
-            const auto num_pages{std::min(
+            const auto min_num_pages{std::min(
                it.GetNumPages(), (chosen_manager.GetEndAddress() - it.GetAddress()) / PageSize)};
-            chosen_manager.Free(it.GetAddress(), num_pages);
+            chosen_manager.Free(it.GetAddress(), min_num_pages);
        }
    });
@@ -165,9 +165,9 @@ ResultCode MemoryManager::Free(PageLinkedList& page_list, std::size_t num_pages,
    // Free all of the pages
    for (const auto& it : page_list.Nodes()) {
-        const auto num_pages{std::min(
+        const auto min_num_pages{std::min(
            it.GetNumPages(), (chosen_manager.GetEndAddress() - it.GetAddress()) / PageSize)};
-        chosen_manager.Free(it.GetAddress(), num_pages);
+        chosen_manager.Free(it.GetAddress(), min_num_pages);
    }
    return RESULT_SUCCESS;
diff --git a/src/core/hle/kernel/memory/memory_manager.h b/src/core/hle/kernel/memory/memory_manager.h
index b078d7a5e..3cf444857 100644
--- a/src/core/hle/kernel/memory/memory_manager.h
+++ b/src/core/hle/kernel/memory/memory_manager.h
@@ -7,7 +7,6 @@
 #include <array>
 #include <mutex>
-#include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "core/hle/kernel/memory/page_heap.h"
 #include "core/hle/result.h"
diff --git a/src/core/hle/kernel/memory/page_heap.cpp b/src/core/hle/kernel/memory/page_heap.cpp
index efcbb3cad..0ab1f7205 100644
--- a/src/core/hle/kernel/memory/page_heap.cpp
+++ b/src/core/hle/kernel/memory/page_heap.cpp
@@ -2,8 +2,8 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-// This file references various implementation details from Atmosph�re, an open-source firmware for
+// This file references various implementation details from Atmosphere, an open-source firmware for
-// the Nintendo Switch. Copyright 2018-2020 Atmosph�re-NX.
+// the Nintendo Switch. Copyright 2018-2020 Atmosphere-NX.
 #include "core/core.h"
 #include "core/hle/kernel/memory/page_heap.h"
diff --git a/src/core/hle/kernel/memory/page_heap.h b/src/core/hle/kernel/memory/page_heap.h
index 380c3f5a1..22b0de860 100644
--- a/src/core/hle/kernel/memory/page_heap.h
+++ b/src/core/hle/kernel/memory/page_heap.h
@@ -2,8 +2,8 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-// This file references various implementation details from Atmosph�re, an open-source firmware for
+// This file references various implementation details from Atmosphere, an open-source firmware for
-// the Nintendo Switch. Copyright 2018-2020 Atmosph�re-NX.
+// the Nintendo Switch. Copyright 2018-2020 Atmosphere-NX.
 #pragma once
diff --git a/src/core/hle/kernel/memory/page_linked_list.h b/src/core/hle/kernel/memory/page_linked_list.h
index 0668d00c6..45dc13eaf 100644
--- a/src/core/hle/kernel/memory/page_linked_list.h
+++ b/src/core/hle/kernel/memory/page_linked_list.h
@@ -7,7 +7,6 @@
 #include <list>
 #include "common/assert.h"
-#include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "core/hle/kernel/memory/memory_types.h"
 #include "core/hle/result.h"
diff --git a/src/core/hle/kernel/memory/page_table.cpp b/src/core/hle/kernel/memory/page_table.cpp
index 3281611f8..5d6aac00f 100644
--- a/src/core/hle/kernel/memory/page_table.cpp
+++ b/src/core/hle/kernel/memory/page_table.cpp
@@ -6,7 +6,6 @@
 #include "common/assert.h"
 #include "common/scope_exit.h"
 #include "core/core.h"
-#include "core/device_memory.h"
 #include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/kernel.h"
 #include "core/hle/kernel/memory/address_space_info.h"
diff --git a/src/core/hle/kernel/memory/page_table.h b/src/core/hle/kernel/memory/page_table.h
index a867aa050..ce0d38849 100644
--- a/src/core/hle/kernel/memory/page_table.h
+++ b/src/core/hle/kernel/memory/page_table.h
@@ -4,16 +4,15 @@
 #pragma once
-#include <list>
 #include <memory>
 #include <mutex>
-#include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "common/page_table.h"
 #include "core/file_sys/program_metadata.h"
 #include "core/hle/kernel/memory/memory_block.h"
 #include "core/hle/kernel/memory/memory_manager.h"
+#include "core/hle/result.h"
 namespace Core {
 class System;
diff --git a/src/core/hle/kernel/memory/slab_heap.h b/src/core/hle/kernel/memory/slab_heap.h
index be95fc3f7..465eaddb3 100644
--- a/src/core/hle/kernel/memory/slab_heap.h
+++ b/src/core/hle/kernel/memory/slab_heap.h
@@ -2,15 +2,14 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-// This file references various implementation details from Atmosph�re, an open-source firmware for
+// This file references various implementation details from Atmosphere, an open-source firmware for
-// the Nintendo Switch. Copyright 2018-2020 Atmosph�re-NX.
+// the Nintendo Switch. Copyright 2018-2020 Atmosphere-NX.
 #pragma once
 #include <atomic>
 #include "common/assert.h"
-#include "common/common_funcs.h"
 #include "common/common_types.h"
 namespace Kernel::Memory {
diff --git a/src/core/hle/kernel/memory/system_control.cpp b/src/core/hle/kernel/memory/system_control.cpp
index 9cae3c6cb..2f98e9c4c 100644
--- a/src/core/hle/kernel/memory/system_control.cpp
+++ b/src/core/hle/kernel/memory/system_control.cpp
@@ -2,8 +2,6 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#pragma once
 #include <random>
 #include "core/hle/kernel/memory/system_control.h"
diff --git a/src/core/hle/kernel/mutex.cpp b/src/core/hle/kernel/mutex.cpp
index eff4e45b0..7869eb32b 100644
--- a/src/core/hle/kernel/mutex.cpp
+++ b/src/core/hle/kernel/mutex.cpp
@@ -7,6 +7,7 @@
 #include <vector>
 #include "common/assert.h"
+#include "common/logging/log.h"
 #include "core/core.h"
 #include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/handle_table.h"
@@ -67,6 +68,7 @@ ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle,
                             Handle requesting_thread_handle) {
    // The mutex address must be 4-byte aligned
    if ((address % sizeof(u32)) != 0) {
+        LOG_ERROR(Kernel, "Address is not 4-byte aligned! address={:016X}", address);
        return ERR_INVALID_ADDRESS;
    }
@@ -88,6 +90,8 @@ ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle,
    }
    if (holding_thread == nullptr) {
+        LOG_ERROR(Kernel, "Holding thread does not exist! thread_handle={:08X}",
+                  holding_thread_handle);
        return ERR_INVALID_HANDLE;
    }
@@ -109,6 +113,7 @@ ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle,
 ResultCode Mutex::Release(VAddr address) {
    // The mutex address must be 4-byte aligned
    if ((address % sizeof(u32)) != 0) {
+        LOG_ERROR(Kernel, "Address is not 4-byte aligned! address={:016X}", address);
        return ERR_INVALID_ADDRESS;
    }
diff --git a/src/core/hle/kernel/process_capability.cpp b/src/core/hle/kernel/process_capability.cpp
index 48e5ae682..63880f13d 100644
--- a/src/core/hle/kernel/process_capability.cpp
+++ b/src/core/hle/kernel/process_capability.cpp
@@ -3,6 +3,7 @@
 // Refer to the license.txt file included.
 #include "common/bit_util.h"
+#include "common/logging/log.h"
 #include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/handle_table.h"
 #include "core/hle/kernel/memory/page_table.h"
@@ -119,22 +120,30 @@ ResultCode ProcessCapabilities::ParseCapabilities(const u32* capabilities,
            // The MapPhysical type uses two descriptor flags for its parameters.
            // If there's only one, then there's a problem.
            if (i >= num_capabilities) {
+                LOG_ERROR(Kernel, "Invalid combination! i={}", i);
                return ERR_INVALID_COMBINATION;
            }
            const auto size_flags = capabilities[i];
            if (GetCapabilityType(size_flags) != CapabilityType::MapPhysical) {
+                LOG_ERROR(Kernel, "Invalid capability type! size_flags={}", size_flags);
                return ERR_INVALID_COMBINATION;
            }
            const auto result = HandleMapPhysicalFlags(descriptor, size_flags, page_table);
            if (result.IsError()) {
+                LOG_ERROR(Kernel, "Failed to map physical flags! descriptor={}, size_flags={}",
+                          descriptor, size_flags);
                return result;
            }
        } else {
            const auto result =
                ParseSingleFlagCapability(set_flags, set_svc_bits, descriptor, page_table);
            if (result.IsError()) {
+                LOG_ERROR(
+                    Kernel,
+                    "Failed to parse capability flag! set_flags={}, set_svc_bits={}, descriptor={}",
+                    set_flags, set_svc_bits, descriptor);
                return result;
            }
        }
@@ -162,6 +171,9 @@ ResultCode ProcessCapabilities::ParseSingleFlagCapability(u32& set_flags, u32& s
    const u32 flag_length = GetFlagBitOffset(type);
    const u32 set_flag = 1U << flag_length;
    if ((set_flag & set_flags & InitializeOnceMask) != 0) {
+        LOG_ERROR(Kernel,
+                  "Attempted to initialize flags that may only be initialized once. set_flags={}",
+                  set_flags);
        return ERR_INVALID_COMBINATION;
    }
    set_flags |= set_flag;
@@ -187,6 +199,7 @@ ResultCode ProcessCapabilities::ParseSingleFlagCapability(u32& set_flags, u32& s
        break;
    }
+    LOG_ERROR(Kernel, "Invalid capability type! type={}", static_cast<u32>(type));
    return ERR_INVALID_CAPABILITY_DESCRIPTOR;
 }
@@ -208,23 +221,31 @@ void ProcessCapabilities::Clear() {
 ResultCode ProcessCapabilities::HandlePriorityCoreNumFlags(u32 flags) {
    if (priority_mask != 0 || core_mask != 0) {
+        LOG_ERROR(Kernel, "Core or priority mask are not zero! priority_mask={}, core_mask={}",
+                  priority_mask, core_mask);
        return ERR_INVALID_CAPABILITY_DESCRIPTOR;
    }
    const u32 core_num_min = (flags >> 16) & 0xFF;
    const u32 core_num_max = (flags >> 24) & 0xFF;
    if (core_num_min > core_num_max) {
+        LOG_ERROR(Kernel, "Core min is greater than core max! core_num_min={}, core_num_max={}",
+                  core_num_min, core_num_max);
        return ERR_INVALID_COMBINATION;
    }
    const u32 priority_min = (flags >> 10) & 0x3F;
    const u32 priority_max = (flags >> 4) & 0x3F;
    if (priority_min > priority_max) {
+        LOG_ERROR(Kernel,
+                  "Priority min is greater than priority max! priority_min={}, priority_max={}",
+                  core_num_min, priority_max);
        return ERR_INVALID_COMBINATION;
    }
    // The switch only has 4 usable cores.
    if (core_num_max >= 4) {
+        LOG_ERROR(Kernel, "Invalid max cores specified! core_num_max={}", core_num_max);
        return ERR_INVALID_PROCESSOR_ID;
    }
@@ -259,6 +280,7 @@ ResultCode ProcessCapabilities::HandleSyscallFlags(u32& set_svc_bits, u32 flags)
        }
        if (svc_number >= svc_capabilities.size()) {
+            LOG_ERROR(Kernel, "Process svc capability is out of range! svc_number={}", svc_number);
            return ERR_OUT_OF_RANGE;
        }
@@ -295,6 +317,8 @@ ResultCode ProcessCapabilities::HandleInterruptFlags(u32 flags) {
        // emulate that, it's sufficient to mark every interrupt as defined.
        if (interrupt >= interrupt_capabilities.size()) {
+            LOG_ERROR(Kernel, "Process interrupt capability is out of range! svc_number={}",
+                      interrupt);
            return ERR_OUT_OF_RANGE;
        }
@@ -307,6 +331,7 @@ ResultCode ProcessCapabilities::HandleInterruptFlags(u32 flags) {
 ResultCode ProcessCapabilities::HandleProgramTypeFlags(u32 flags) {
    const u32 reserved = flags >> 17;
    if (reserved != 0) {
+        LOG_ERROR(Kernel, "Reserved value is non-zero! reserved={}", reserved);
        return ERR_RESERVED_VALUE;
    }
@@ -324,6 +349,9 @@ ResultCode ProcessCapabilities::HandleKernelVersionFlags(u32 flags) {
    const u32 major_version = kernel_version >> 19;
    if (major_version != 0 || flags < 0x80000) {
+        LOG_ERROR(Kernel,
+                  "Kernel version is non zero or flags are too small! major_version={}, flags={}",
+                  major_version, flags);
        return ERR_INVALID_CAPABILITY_DESCRIPTOR;
    }
@@ -334,6 +362,7 @@ ResultCode ProcessCapabilities::HandleKernelVersionFlags(u32 flags) {
 ResultCode ProcessCapabilities::HandleHandleTableFlags(u32 flags) {
    const u32 reserved = flags >> 26;
    if (reserved != 0) {
+        LOG_ERROR(Kernel, "Reserved value is non-zero! reserved={}", reserved);
        return ERR_RESERVED_VALUE;
    }
@@ -344,6 +373,7 @@ ResultCode ProcessCapabilities::HandleHandleTableFlags(u32 flags) {
 ResultCode ProcessCapabilities::HandleDebugFlags(u32 flags) {
    const u32 reserved = flags >> 19;
    if (reserved != 0) {
+        LOG_ERROR(Kernel, "Reserved value is non-zero! reserved={}", reserved);
        return ERR_RESERVED_VALUE;
    }
diff --git a/src/core/hle/kernel/readable_event.cpp b/src/core/hle/kernel/readable_event.cpp
index 9d3d3a81b..00860fcbd 100644
--- a/src/core/hle/kernel/readable_event.cpp
+++ b/src/core/hle/kernel/readable_event.cpp
@@ -4,6 +4,7 @@
 #include <algorithm>
 #include "common/assert.h"
+#include "common/logging/log.h"
 #include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/object.h"
 #include "core/hle/kernel/readable_event.h"
@@ -23,10 +24,12 @@ void ReadableEvent::Acquire(Thread* thread) {
 }
 void ReadableEvent::Signal() {
-    if (!is_signaled) {
+    if (is_signaled) {
-        is_signaled = true;
+        return;
-        SynchronizationObject::Signal();
+    }
-    };
+    is_signaled = true;
+    SynchronizationObject::Signal();
 }
 void ReadableEvent::Clear() {
@@ -35,6 +38,8 @@ void ReadableEvent::Clear() {
 ResultCode ReadableEvent::Reset() {
    if (!is_signaled) {
+        LOG_ERROR(Kernel, "Handle is not signaled! object_id={}, object_type={}, object_name={}",
+                  GetObjectId(), GetTypeName(), GetName());
        return ERR_INVALID_STATE;
    }
diff --git a/src/core/hle/kernel/resource_limit.cpp b/src/core/hle/kernel/resource_limit.cpp
index 96e5b9892..d9beaa3a4 100644
--- a/src/core/hle/kernel/resource_limit.cpp
+++ b/src/core/hle/kernel/resource_limit.cpp
@@ -69,6 +69,8 @@ ResultCode ResourceLimit::SetLimitValue(ResourceType resource, s64 value) {
        limit[index] = value;
        return RESULT_SUCCESS;
    } else {
+        LOG_ERROR(Kernel, "Limit value is too large! resource={}, value={}, index={}",
+                  static_cast<u32>(resource), value, index);
        return ERR_INVALID_STATE;
    }
 }
diff --git a/src/core/hle/kernel/server_session.cpp b/src/core/hle/kernel/server_session.cpp
index 0f102ca44..25438b86b 100644
--- a/src/core/hle/kernel/server_session.cpp
+++ b/src/core/hle/kernel/server_session.cpp
@@ -137,8 +137,8 @@ ResultCode ServerSession::HandleDomainSyncRequest(Kernel::HLERequestContext& con
 ResultCode ServerSession::QueueSyncRequest(std::shared_ptr<Thread> thread,
                                           Core::Memory::Memory& memory) {
    u32* cmd_buf{reinterpret_cast<u32*>(memory.GetPointer(thread->GetTLSAddress()))};
-    std::shared_ptr<Kernel::HLERequestContext> context{
+    auto context =
-        std::make_shared<Kernel::HLERequestContext>(SharedFrom(this), std::move(thread))};
+        std::make_shared<HLERequestContext>(kernel, memory, SharedFrom(this), std::move(thread));
    context->PopulateFromIncomingCommandBuffer(kernel.CurrentProcess()->GetHandleTable(), cmd_buf);
    request_queue.Push(std::move(context));
diff --git a/src/core/hle/kernel/svc.cpp b/src/core/hle/kernel/svc.cpp
index 41ef2caf6..4ae4529f5 100644
--- a/src/core/hle/kernel/svc.cpp
+++ b/src/core/hle/kernel/svc.cpp
@@ -685,6 +685,8 @@ static ResultCode GetInfo(Core::System& system, u64* result, u64 info_id, u64 ha
    case GetInfoType::TotalPhysicalMemoryAvailableWithoutSystemResource:
    case GetInfoType::TotalPhysicalMemoryUsedWithoutSystemResource: {
        if (info_sub_id != 0) {
+            LOG_ERROR(Kernel_SVC, "Info sub id is non zero! info_id={}, info_sub_id={}", info_id,
+                      info_sub_id);
            return ERR_INVALID_ENUM_VALUE;
        }
@@ -692,6 +694,8 @@ static ResultCode GetInfo(Core::System& system, u64* result, u64 info_id, u64 ha
            system.Kernel().CurrentProcess()->GetHandleTable();
        const auto process = current_process_handle_table.Get<Process>(static_cast<Handle>(handle));
        if (!process) {
+            LOG_ERROR(Kernel_SVC, "Process is not valid! info_id={}, info_sub_id={}, handle={:08X}",
+                      info_id, info_sub_id, handle);
            return ERR_INVALID_HANDLE;
        }
@@ -783,10 +787,13 @@ static ResultCode GetInfo(Core::System& system, u64* result, u64 info_id, u64 ha
    case GetInfoType::RegisterResourceLimit: {
        if (handle != 0) {
+            LOG_ERROR(Kernel, "Handle is non zero! handle={:08X}", handle);
            return ERR_INVALID_HANDLE;
        }
        if (info_sub_id != 0) {
+            LOG_ERROR(Kernel, "Info sub id is non zero! info_id={}, info_sub_id={}", info_id,
+                      info_sub_id);
            return ERR_INVALID_COMBINATION;
        }
diff --git a/src/core/hle/kernel/thread.cpp b/src/core/hle/kernel/thread.cpp
index a919750a6..db7f379ac 100644
--- a/src/core/hle/kernel/thread.cpp
+++ b/src/core/hle/kernel/thread.cpp
@@ -423,6 +423,8 @@ ResultCode Thread::SetCoreAndAffinityMask(s32 new_core, u64 new_affinity_mask) {
    if (new_core == THREADPROCESSORID_DONT_UPDATE) {
        new_core = use_override ? ideal_core_override : ideal_core;
        if ((new_affinity_mask & (1ULL << new_core)) == 0) {
+            LOG_ERROR(Kernel, "New affinity mask is incorrect! new_core={}, new_affinity_mask={}",
+                      new_core, new_affinity_mask);
            return ERR_INVALID_COMBINATION;
        }
    }
diff --git a/src/core/hle/service/acc/acc.cpp b/src/core/hle/service/acc/acc.cpp
index 9a7992f58..630a8b048 100644
--- a/src/core/hle/service/acc/acc.cpp
+++ b/src/core/hle/service/acc/acc.cpp
@@ -228,7 +228,8 @@ public:
 class IManagerForApplication final : public ServiceFramework<IManagerForApplication> {
 public:
-    IManagerForApplication() : ServiceFramework("IManagerForApplication") {
+    explicit IManagerForApplication(Common::UUID user_id)
+        : ServiceFramework("IManagerForApplication"), user_id(user_id) {
        // clang-format off
        static const FunctionInfo functions[] = {
            {0, &IManagerForApplication::CheckAvailability, "CheckAvailability"},
@@ -254,12 +255,14 @@ private:
    }
    void GetAccountId(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_ACC, "(STUBBED) called");
+        LOG_DEBUG(Service_ACC, "called");
-        // Should return a nintendo account ID
        IPC::ResponseBuilder rb{ctx, 4};
        rb.Push(RESULT_SUCCESS);
-        rb.PushRaw<u64>(1);
+        rb.PushRaw<u64>(user_id.GetNintendoID());
    }
+    Common::UUID user_id;
 };
 void Module::Interface::GetUserCount(Kernel::HLERequestContext& ctx) {
@@ -382,7 +385,7 @@ void Module::Interface::GetBaasAccountManagerForApplication(Kernel::HLERequestCo
    LOG_DEBUG(Service_ACC, "called");
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
-    rb.PushIpcInterface<IManagerForApplication>();
+    rb.PushIpcInterface<IManagerForApplication>(profile_manager->GetLastOpenedUser());
 }
 void Module::Interface::IsUserAccountSwitchLocked(Kernel::HLERequestContext& ctx) {
diff --git a/src/core/hle/service/am/am.cpp b/src/core/hle/service/am/am.cpp
index 9450de6e9..4df74c4f9 100644
--- a/src/core/hle/service/am/am.cpp
+++ b/src/core/hle/service/am/am.cpp
@@ -1335,12 +1335,23 @@ void IApplicationFunctions::SetTerminateResult(Kernel::HLERequestContext& ctx) {
 }
 void IApplicationFunctions::GetDisplayVersion(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    LOG_DEBUG(Service_AM, "called");
+    std::array<u8, 0x10> version_string{};
+    FileSys::PatchManager pm{system.CurrentProcess()->GetTitleID()};
+    const auto res = pm.GetControlMetadata();
+    if (res.first != nullptr) {
+        const auto& version = res.first->GetVersionString();
+        std::copy(version.begin(), version.end(), version_string.begin());
+    } else {
+        constexpr u128 default_version = {1, 0};
+        std::memcpy(version_string.data(), default_version.data(), sizeof(u128));
+    }
    IPC::ResponseBuilder rb{ctx, 6};
    rb.Push(RESULT_SUCCESS);
-    rb.Push<u64>(1);
+    rb.PushRaw(version_string);
-    rb.Push<u64>(0);
 }
 void IApplicationFunctions::GetDesiredLanguage(Kernel::HLERequestContext& ctx) {
@@ -1514,14 +1525,15 @@ void InstallInterfaces(SM::ServiceManager& service_manager,
    std::make_shared<TCAP>()->InstallAsService(service_manager);
 }
-IHomeMenuFunctions::IHomeMenuFunctions() : ServiceFramework("IHomeMenuFunctions") {
+IHomeMenuFunctions::IHomeMenuFunctions(Kernel::KernelCore& kernel)
+    : ServiceFramework("IHomeMenuFunctions"), kernel(kernel) {
    // clang-format off
    static const FunctionInfo functions[] = {
        {10, &IHomeMenuFunctions::RequestToGetForeground, "RequestToGetForeground"},
        {11, nullptr, "LockForeground"},
        {12, nullptr, "UnlockForeground"},
        {20, nullptr, "PopFromGeneralChannel"},
-        {21, nullptr, "GetPopFromGeneralChannelEvent"},
+        {21, &IHomeMenuFunctions::GetPopFromGeneralChannelEvent, "GetPopFromGeneralChannelEvent"},
        {30, nullptr, "GetHomeButtonWriterLockAccessor"},
        {31, nullptr, "GetWriterLockAccessorEx"},
        {100, nullptr, "PopRequestLaunchApplicationForDebug"},
@@ -1531,6 +1543,9 @@ IHomeMenuFunctions::IHomeMenuFunctions() : ServiceFramework("IHomeMenuFunctions"
    // clang-format on
    RegisterHandlers(functions);
+    pop_from_general_channel_event = Kernel::WritableEvent::CreateEventPair(
+        kernel, "IHomeMenuFunctions:PopFromGeneralChannelEvent");
 }
 IHomeMenuFunctions::~IHomeMenuFunctions() = default;
@@ -1542,6 +1557,14 @@ void IHomeMenuFunctions::RequestToGetForeground(Kernel::HLERequestContext& ctx)
    rb.Push(RESULT_SUCCESS);
 }
+void IHomeMenuFunctions::GetPopFromGeneralChannelEvent(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_AM, "(STUBBED) called");
+    IPC::ResponseBuilder rb{ctx, 2, 1};
+    rb.Push(RESULT_SUCCESS);
+    rb.PushCopyObjects(pop_from_general_channel_event.readable);
+}
 IGlobalStateController::IGlobalStateController() : ServiceFramework("IGlobalStateController") {
    // clang-format off
    static const FunctionInfo functions[] = {
diff --git a/src/core/hle/service/am/am.h b/src/core/hle/service/am/am.h
index dfa701d73..469f7f814 100644
--- a/src/core/hle/service/am/am.h
+++ b/src/core/hle/service/am/am.h
@@ -292,11 +292,15 @@ private:
 class IHomeMenuFunctions final : public ServiceFramework<IHomeMenuFunctions> {
 public:
-    IHomeMenuFunctions();
+    explicit IHomeMenuFunctions(Kernel::KernelCore& kernel);
    ~IHomeMenuFunctions() override;
 private:
    void RequestToGetForeground(Kernel::HLERequestContext& ctx);
+    void GetPopFromGeneralChannelEvent(Kernel::HLERequestContext& ctx);
+    Kernel::EventPair pop_from_general_channel_event;
+    Kernel::KernelCore& kernel;
 };
 class IGlobalStateController final : public ServiceFramework<IGlobalStateController> {
diff --git a/src/core/hle/service/am/applet_ae.cpp b/src/core/hle/service/am/applet_ae.cpp
index e454b77d8..9df286d17 100644
--- a/src/core/hle/service/am/applet_ae.cpp
+++ b/src/core/hle/service/am/applet_ae.cpp
@@ -202,7 +202,7 @@ private:
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
-        rb.PushIpcInterface<IHomeMenuFunctions>();
+        rb.PushIpcInterface<IHomeMenuFunctions>(system.Kernel());
    }
    void GetGlobalStateController(Kernel::HLERequestContext& ctx) {
diff --git a/src/core/hle/service/audio/audin_u.cpp b/src/core/hle/service/audio/audin_u.cpp
index d7f1d348d..3e2299426 100644
--- a/src/core/hle/service/audio/audin_u.cpp
+++ b/src/core/hle/service/audio/audin_u.cpp
@@ -2,6 +2,9 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
+#include "common/logging/log.h"
+#include "core/hle/ipc_helpers.h"
+#include "core/hle/kernel/hle_ipc.h"
 #include "core/hle/service/audio/audin_u.h"
 namespace Service::Audio {
@@ -36,11 +39,12 @@ public:
 AudInU::AudInU() : ServiceFramework("audin:u") {
    // clang-format off
    static const FunctionInfo functions[] = {
-        {0, nullptr, "ListAudioIns"},
+        {0, &AudInU::ListAudioIns, "ListAudioIns"},
-        {1, nullptr, "OpenAudioIn"},
+        {1, &AudInU::OpenAudioIn, "OpenAudioIn"},
-        {2, nullptr, "Unknown"},
+        {2, &AudInU::ListAudioIns, "ListAudioInsAuto"},
-        {3, nullptr, "OpenAudioInAuto"},
+        {3, &AudInU::OpenAudioIn, "OpenAudioInAuto"},
-        {4, nullptr, "ListAudioInsAuto"},
+        {4, &AudInU::ListAudioInsAutoFiltered, "ListAudioInsAutoFiltered"},
+        {5, &AudInU::OpenAudioInProtocolSpecified, "OpenAudioInProtocolSpecified"},
    };
    // clang-format on
@@ -49,4 +53,60 @@ AudInU::AudInU() : ServiceFramework("audin:u") {
 AudInU::~AudInU() = default;
+void AudInU::ListAudioIns(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Audio, "called");
+    const std::size_t count = ctx.GetWriteBufferSize() / sizeof(AudioInDeviceName);
+    const std::size_t device_count = std::min(count, audio_device_names.size());
+    std::vector<AudioInDeviceName> device_names;
+    device_names.reserve(device_count);
+    for (std::size_t i = 0; i < device_count; i++) {
+        const auto& device_name = audio_device_names[i];
+        auto& entry = device_names.emplace_back();
+        device_name.copy(entry.data(), device_name.size());
+    }
+    ctx.WriteBuffer(device_names);
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(RESULT_SUCCESS);
+    rb.Push(static_cast<u32>(device_names.size()));
+}
+void AudInU::ListAudioInsAutoFiltered(Kernel::HLERequestContext& ctx) {
+    LOG_DEBUG(Service_Audio, "called");
+    constexpr u32 device_count = 0;
+    // Since we don't actually use any other audio input devices, we return 0 devices. Filtered
+    // device listing just omits the default input device
+    IPC::ResponseBuilder rb{ctx, 3};
+    rb.Push(RESULT_SUCCESS);
+    rb.Push(static_cast<u32>(device_count));
+}
+void AudInU::OpenInOutImpl(Kernel::HLERequestContext& ctx) {
+    AudInOutParams params{};
+    params.channel_count = 2;
+    params.sample_format = SampleFormat::PCM16;
+    params.sample_rate = 48000;
+    params.state = State::Started;
+    IPC::ResponseBuilder rb{ctx, 6, 0, 1};
+    rb.Push(RESULT_SUCCESS);
+    rb.PushRaw<AudInOutParams>(params);
+    rb.PushIpcInterface<IAudioIn>();
+}
+void AudInU::OpenAudioIn(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_Audio, "(STUBBED) called");
+    OpenInOutImpl(ctx);
+}
+void AudInU::OpenAudioInProtocolSpecified(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_Audio, "(STUBBED) called");
+    OpenInOutImpl(ctx);
+}
 } // namespace Service::Audio
diff --git a/src/core/hle/service/audio/audin_u.h b/src/core/hle/service/audio/audin_u.h
index 0538b9560..a599f4a64 100644
--- a/src/core/hle/service/audio/audin_u.h
+++ b/src/core/hle/service/audio/audin_u.h
@@ -16,6 +16,35 @@ class AudInU final : public ServiceFramework<AudInU> {
 public:
    explicit AudInU();
    ~AudInU() override;
+private:
+    enum class SampleFormat : u32_le {
+        PCM16 = 2,
+    };
+    enum class State : u32_le {
+        Started = 0,
+        Stopped = 1,
+    };
+    struct AudInOutParams {
+        u32_le sample_rate{};
+        u32_le channel_count{};
+        SampleFormat sample_format{};
+        State state{};
+    };
+    static_assert(sizeof(AudInOutParams) == 0x10, "AudInOutParams is an invalid size");
+    using AudioInDeviceName = std::array<char, 256>;
+    static constexpr std::array<std::string_view, 1> audio_device_names{{
+        "BuiltInHeadset",
+    }};
+    void ListAudioIns(Kernel::HLERequestContext& ctx);
+    void ListAudioInsAutoFiltered(Kernel::HLERequestContext& ctx);
+    void OpenInOutImpl(Kernel::HLERequestContext& ctx);
+    void OpenAudioIn(Kernel::HLERequestContext& ctx);
+    void OpenAudioInProtocolSpecified(Kernel::HLERequestContext& ctx);
 };
 } // namespace Service::Audio
diff --git a/src/core/hle/service/bcat/backend/boxcat.cpp b/src/core/hle/service/bcat/backend/boxcat.cpp
index 5febe8fc1..d29e78d7e 100644
--- a/src/core/hle/service/bcat/backend/boxcat.cpp
+++ b/src/core/hle/service/bcat/backend/boxcat.cpp
@@ -4,8 +4,8 @@
 #include <fmt/ostream.h>
 #include <httplib.h>
-#include <json.hpp>
 #include <mbedtls/sha256.h>
+#include <nlohmann/json.hpp>
 #include "common/hex_util.h"
 #include "common/logging/backend.h"
 #include "common/logging/log.h"
diff --git a/src/core/hle/service/hid/hid.cpp b/src/core/hle/service/hid/hid.cpp
index 5559587e3..c84cb1483 100644
--- a/src/core/hle/service/hid/hid.cpp
+++ b/src/core/hle/service/hid/hid.cpp
@@ -157,7 +157,7 @@ Hid::Hid(Core::System& system) : ServiceFramework("hid"), system(system) {
        {11, &Hid::ActivateTouchScreen, "ActivateTouchScreen"},
        {21, &Hid::ActivateMouse, "ActivateMouse"},
        {31, &Hid::ActivateKeyboard, "ActivateKeyboard"},
-        {32, nullptr, "SendKeyboardLockKeyEvent"},
+        {32, &Hid::SendKeyboardLockKeyEvent, "SendKeyboardLockKeyEvent"},
        {40, nullptr, "AcquireXpadIdEventHandle"},
        {41, nullptr, "ReleaseXpadIdEventHandle"},
        {51, &Hid::ActivateXpad, "ActivateXpad"},
@@ -871,6 +871,15 @@ void Hid::InitializeSevenSixAxisSensor(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
 }
+void Hid::SendKeyboardLockKeyEvent(Kernel::HLERequestContext& ctx) {
+    IPC::RequestParser rp{ctx};
+    const auto flags{rp.Pop<u32>()};
+    LOG_WARNING(Service_HID, "(STUBBED) called. flags={}", flags);
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
 class HidDbg final : public ServiceFramework<HidDbg> {
 public:
    explicit HidDbg() : ServiceFramework{"hid:dbg"} {
diff --git a/src/core/hle/service/hid/hid.h b/src/core/hle/service/hid/hid.h
index 23552efb1..c8ed4ad8b 100644
--- a/src/core/hle/service/hid/hid.h
+++ b/src/core/hle/service/hid/hid.h
@@ -130,6 +130,7 @@ private:
    void SetPalmaBoostMode(Kernel::HLERequestContext& ctx);
    void StopSevenSixAxisSensor(Kernel::HLERequestContext& ctx);
    void InitializeSevenSixAxisSensor(Kernel::HLERequestContext& ctx);
+    void SendKeyboardLockKeyEvent(Kernel::HLERequestContext& ctx);
    std::shared_ptr<IAppletResource> applet_resource;
    Core::System& system;
diff --git a/src/core/reporter.cpp b/src/core/reporter.cpp
index 558cbe6d7..76cfa5a17 100644
--- a/src/core/reporter.cpp
+++ b/src/core/reporter.cpp
@@ -4,11 +4,12 @@
 #include <ctime>
 #include <fstream>
+#include <iomanip>
 #include <fmt/chrono.h>
 #include <fmt/format.h>
 #include <fmt/ostream.h>
-#include <json.hpp>
+#include <nlohmann/json.hpp>
 #include "common/file_util.h"
 #include "common/hex_util.h"
diff --git a/src/core/settings.cpp b/src/core/settings.cpp
index cd6c257f5..2b0bdc4d3 100644
--- a/src/core/settings.cpp
+++ b/src/core/settings.cpp
@@ -96,6 +96,7 @@ void LogSettings() {
    LogSetting("Renderer_UseAsynchronousGpuEmulation",
               Settings::values.use_asynchronous_gpu_emulation);
    LogSetting("Renderer_UseVsync", Settings::values.use_vsync);
+    LogSetting("Renderer_AnisotropicFilteringLevel", Settings::values.max_anisotropy);
    LogSetting("Audio_OutputEngine", Settings::values.sink_id);
    LogSetting("Audio_EnableAudioStretching", Settings::values.enable_audio_stretching);
    LogSetting("Audio_OutputDevice", Settings::values.audio_device_id);
diff --git a/src/input_common/main.cpp b/src/input_common/main.cpp
index c98c848cf..95e351e24 100644
--- a/src/input_common/main.cpp
+++ b/src/input_common/main.cpp
@@ -18,7 +18,9 @@ namespace InputCommon {
 static std::shared_ptr<Keyboard> keyboard;
 static std::shared_ptr<MotionEmu> motion_emu;
+#ifdef HAVE_SDL2
 static std::unique_ptr<SDL::State> sdl;
+#endif
 static std::unique_ptr<CemuhookUDP::State> udp;
 void Init() {
@@ -29,7 +31,9 @@ void Init() {
    motion_emu = std::make_shared<MotionEmu>();
    Input::RegisterFactory<Input::MotionDevice>("motion_emu", motion_emu);
+#ifdef HAVE_SDL2
    sdl = SDL::Init();
+#endif
    udp = CemuhookUDP::Init();
 }
@@ -40,7 +44,9 @@ void Shutdown() {
    Input::UnregisterFactory<Input::AnalogDevice>("analog_from_button");
    Input::UnregisterFactory<Input::MotionDevice>("motion_emu");
    motion_emu.reset();
+#ifdef HAVE_SDL2
    sdl.reset();
+#endif
    udp.reset();
 }
diff --git a/src/video_core/CMakeLists.txt b/src/video_core/CMakeLists.txt
index ff53282c9..d23c53843 100644
--- a/src/video_core/CMakeLists.txt
+++ b/src/video_core/CMakeLists.txt
@@ -8,6 +8,7 @@ add_library(video_core STATIC
    dma_pusher.h
    engines/const_buffer_engine_interface.h
    engines/const_buffer_info.h
+    engines/engine_interface.h
    engines/engine_upload.cpp
    engines/engine_upload.h
    engines/fermi_2d.cpp
diff --git a/src/video_core/buffer_cache/buffer_cache.h b/src/video_core/buffer_cache/buffer_cache.h
index 510f11089..56e570994 100644
--- a/src/video_core/buffer_cache/buffer_cache.h
+++ b/src/video_core/buffer_cache/buffer_cache.h
@@ -4,7 +4,6 @@
 #pragma once
-#include <array>
 #include <list>
 #include <memory>
 #include <mutex>
@@ -89,10 +88,6 @@ public:
                map->MarkAsWritten(true);
                MarkRegionAsWritten(map->GetStart(), map->GetEnd() - 1);
            }
-        } else {
-            if (map->IsWritten()) {
-                WriteBarrier();
-            }
        }
        return {ToHandle(block), static_cast<u64>(block->GetOffset(cpu_addr))};
@@ -254,8 +249,6 @@ protected:
    virtual BufferType ToHandle(const OwnerBuffer& storage) = 0;
-    virtual void WriteBarrier() = 0;
    virtual OwnerBuffer CreateBlock(VAddr cpu_addr, std::size_t size) = 0;
    virtual void UploadBlockData(const OwnerBuffer& buffer, std::size_t offset, std::size_t size,
diff --git a/src/video_core/dma_pusher.cpp b/src/video_core/dma_pusher.cpp
index 16311f05e..bdc023d54 100644
--- a/src/video_core/dma_pusher.cpp
+++ b/src/video_core/dma_pusher.cpp
@@ -27,6 +27,8 @@ void DmaPusher::DispatchCalls() {
    dma_pushbuffer_subindex = 0;
+    dma_state.is_last_call = true;
    while (system.IsPoweredOn()) {
        if (!Step()) {
            break;
@@ -82,9 +84,11 @@ bool DmaPusher::Step() {
                    index);
                CallMultiMethod(&command_header.argument, max_write);
                dma_state.method_count -= max_write;
+                dma_state.is_last_call = true;
                index += max_write;
                continue;
            } else {
+                dma_state.is_last_call = dma_state.method_count <= 1;
                CallMethod(command_header.argument);
            }
@@ -144,12 +148,22 @@ void DmaPusher::SetState(const CommandHeader& command_header) {
 }
 void DmaPusher::CallMethod(u32 argument) const {
-    gpu.CallMethod({dma_state.method, argument, dma_state.subchannel, dma_state.method_count});
+    if (dma_state.method < non_puller_methods) {
+        gpu.CallMethod({dma_state.method, argument, dma_state.subchannel, dma_state.method_count});
+    } else {
+        subchannels[dma_state.subchannel]->CallMethod(dma_state.method, argument,
+                                                      dma_state.is_last_call);
+    }
 }
 void DmaPusher::CallMultiMethod(const u32* base_start, u32 num_methods) const {
-    gpu.CallMultiMethod(dma_state.method, dma_state.subchannel, base_start, num_methods,
+    if (dma_state.method < non_puller_methods) {
-                        dma_state.method_count);
+        gpu.CallMultiMethod(dma_state.method, dma_state.subchannel, base_start, num_methods,
+                            dma_state.method_count);
+    } else {
+        subchannels[dma_state.subchannel]->CallMultiMethod(dma_state.method, base_start,
+                                                           num_methods, dma_state.method_count);
+    }
 }
 } // namespace Tegra
diff --git a/src/video_core/dma_pusher.h b/src/video_core/dma_pusher.h
index 6cef71306..e8b714e94 100644
--- a/src/video_core/dma_pusher.h
+++ b/src/video_core/dma_pusher.h
@@ -4,11 +4,13 @@
 #pragma once
+#include <array>
 #include <vector>
 #include <queue>
 #include "common/bit_field.h"
 #include "common/common_types.h"
+#include "video_core/engines/engine_interface.h"
 namespace Core {
 class System;
@@ -69,7 +71,13 @@ public:
    void DispatchCalls();
+    void BindSubchannel(Tegra::Engines::EngineInterface* engine, u32 subchannel_id) {
+        subchannels[subchannel_id] = engine;
+    }
 private:
+    static constexpr u32 non_puller_methods = 0x40;
+    static constexpr u32 max_subchannels = 8;
    bool Step();
    void SetState(const CommandHeader& command_header);
@@ -88,6 +96,7 @@ private:
        u32 method_count;      ///< Current method count
        u32 length_pending;    ///< Large NI command length pending
        bool non_incrementing; ///< Current command's NI flag
+        bool is_last_call;
    };
    DmaState dma_state{};
@@ -96,6 +105,8 @@ private:
    GPUVAddr dma_mget{};  ///< main pushbuffer last read address
    bool ib_enable{true}; ///< IB mode enabled
+    std::array<Tegra::Engines::EngineInterface*, max_subchannels> subchannels{};
    GPU& gpu;
    Core::System& system;
 };
diff --git a/src/video_core/engines/engine_interface.h b/src/video_core/engines/engine_interface.h
new file mode 100644
index 000000000..18a9db7e6
--- /dev/null
+++ b/src/video_core/engines/engine_interface.h
@@ -0,0 +1,22 @@
+// Copyright 2020 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+#pragma once
+#include <type_traits>
+#include "common/common_types.h"
+namespace Tegra::Engines {
+class EngineInterface {
+public:
+    /// Write the value to the register identified by method.
+    virtual void CallMethod(u32 method, u32 method_argument, bool is_last_call) = 0;
+    /// Write multiple values to the register identified by method.
+    virtual void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+                                 u32 methods_pending) = 0;
+};
+} // namespace Tegra::Engines
diff --git a/src/video_core/engines/fermi_2d.cpp b/src/video_core/engines/fermi_2d.cpp
index 8a47614d2..ff10ff40d 100644
--- a/src/video_core/engines/fermi_2d.cpp
+++ b/src/video_core/engines/fermi_2d.cpp
@@ -12,13 +12,13 @@ namespace Tegra::Engines {
 Fermi2D::Fermi2D(VideoCore::RasterizerInterface& rasterizer) : rasterizer{rasterizer} {}
-void Fermi2D::CallMethod(const GPU::MethodCall& method_call) {
+void Fermi2D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS,
               "Invalid Fermi2D register, increase the size of the Regs structure");
-    regs.reg_array[method_call.method] = method_call.argument;
+    regs.reg_array[method] = method_argument;
-    switch (method_call.method) {
+    switch (method) {
    // Trigger the surface copy on the last register write. This is blit_src_y, but this is 64-bit,
    // so trigger on the second 32-bit write.
    case FERMI2D_REG_INDEX(blit_src_y) + 1: {
@@ -30,7 +30,7 @@ void Fermi2D::CallMethod(const GPU::MethodCall& method_call) {
 void Fermi2D::CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending) {
    for (std::size_t i = 0; i < amount; i++) {
-        CallMethod({method, base_start[i], 0, methods_pending - static_cast<u32>(i)});
+        CallMethod(method, base_start[i], methods_pending - static_cast<u32>(i) <= 1);
    }
 }
diff --git a/src/video_core/engines/fermi_2d.h b/src/video_core/engines/fermi_2d.h
index 939a5966d..8f37d053f 100644
--- a/src/video_core/engines/fermi_2d.h
+++ b/src/video_core/engines/fermi_2d.h
@@ -10,6 +10,7 @@
 #include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "common/math_util.h"
+#include "video_core/engines/engine_interface.h"
 #include "video_core/gpu.h"
 namespace Tegra {
@@ -31,16 +32,17 @@ namespace Tegra::Engines {
 #define FERMI2D_REG_INDEX(field_name)                                                              \
    (offsetof(Tegra::Engines::Fermi2D::Regs, field_name) / sizeof(u32))
-class Fermi2D final {
+class Fermi2D final : public EngineInterface {
 public:
    explicit Fermi2D(VideoCore::RasterizerInterface& rasterizer);
    ~Fermi2D() = default;
    /// Write the value to the register identified by method.
-    void CallMethod(const GPU::MethodCall& method_call);
+    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
    /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending);
+    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+                         u32 methods_pending) override;
    enum class Origin : u32 {
        Center = 0,
diff --git a/src/video_core/engines/kepler_compute.cpp b/src/video_core/engines/kepler_compute.cpp
index 00a12175f..f6237fc6a 100644
--- a/src/video_core/engines/kepler_compute.cpp
+++ b/src/video_core/engines/kepler_compute.cpp
@@ -24,20 +24,19 @@ KeplerCompute::KeplerCompute(Core::System& system, VideoCore::RasterizerInterfac
 KeplerCompute::~KeplerCompute() = default;
-void KeplerCompute::CallMethod(const GPU::MethodCall& method_call) {
+void KeplerCompute::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS,
               "Invalid KeplerCompute register, increase the size of the Regs structure");
-    regs.reg_array[method_call.method] = method_call.argument;
+    regs.reg_array[method] = method_argument;
-    switch (method_call.method) {
+    switch (method) {
    case KEPLER_COMPUTE_REG_INDEX(exec_upload): {
        upload_state.ProcessExec(regs.exec_upload.linear != 0);
        break;
    }
    case KEPLER_COMPUTE_REG_INDEX(data_upload): {
-        const bool is_last_call = method_call.IsLastCall();
+        upload_state.ProcessData(method_argument, is_last_call);
-        upload_state.ProcessData(method_call.argument, is_last_call);
        if (is_last_call) {
            system.GPU().Maxwell3D().OnMemoryWrite();
        }
@@ -54,7 +53,7 @@ void KeplerCompute::CallMethod(const GPU::MethodCall& method_call) {
 void KeplerCompute::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
                                    u32 methods_pending) {
    for (std::size_t i = 0; i < amount; i++) {
-        CallMethod({method, base_start[i], 0, methods_pending - static_cast<u32>(i)});
+        CallMethod(method, base_start[i], methods_pending - static_cast<u32>(i) <= 1);
    }
 }
diff --git a/src/video_core/engines/kepler_compute.h b/src/video_core/engines/kepler_compute.h
index fe55fdfd0..18ceedfaf 100644
--- a/src/video_core/engines/kepler_compute.h
+++ b/src/video_core/engines/kepler_compute.h
@@ -11,6 +11,7 @@
 #include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "video_core/engines/const_buffer_engine_interface.h"
+#include "video_core/engines/engine_interface.h"
 #include "video_core/engines/engine_upload.h"
 #include "video_core/engines/shader_type.h"
 #include "video_core/gpu.h"
@@ -39,7 +40,7 @@ namespace Tegra::Engines {
 #define KEPLER_COMPUTE_REG_INDEX(field_name)                                                       \
    (offsetof(Tegra::Engines::KeplerCompute::Regs, field_name) / sizeof(u32))
-class KeplerCompute final : public ConstBufferEngineInterface {
+class KeplerCompute final : public ConstBufferEngineInterface, public EngineInterface {
 public:
    explicit KeplerCompute(Core::System& system, VideoCore::RasterizerInterface& rasterizer,
                           MemoryManager& memory_manager);
@@ -200,10 +201,11 @@ public:
                  "KeplerCompute LaunchParams has wrong size");
    /// Write the value to the register identified by method.
-    void CallMethod(const GPU::MethodCall& method_call);
+    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
    /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending);
+    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+                         u32 methods_pending) override;
    Texture::FullTextureInfo GetTexture(std::size_t offset) const;
diff --git a/src/video_core/engines/kepler_memory.cpp b/src/video_core/engines/kepler_memory.cpp
index 586ff15dc..dc71b2eec 100644
--- a/src/video_core/engines/kepler_memory.cpp
+++ b/src/video_core/engines/kepler_memory.cpp
@@ -19,20 +19,19 @@ KeplerMemory::KeplerMemory(Core::System& system, MemoryManager& memory_manager)
 KeplerMemory::~KeplerMemory() = default;
-void KeplerMemory::CallMethod(const GPU::MethodCall& method_call) {
+void KeplerMemory::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS,
               "Invalid KeplerMemory register, increase the size of the Regs structure");
-    regs.reg_array[method_call.method] = method_call.argument;
+    regs.reg_array[method] = method_argument;
-    switch (method_call.method) {
+    switch (method) {
    case KEPLERMEMORY_REG_INDEX(exec): {
        upload_state.ProcessExec(regs.exec.linear != 0);
        break;
    }
    case KEPLERMEMORY_REG_INDEX(data): {
-        const bool is_last_call = method_call.IsLastCall();
+        upload_state.ProcessData(method_argument, is_last_call);
-        upload_state.ProcessData(method_call.argument, is_last_call);
        if (is_last_call) {
            system.GPU().Maxwell3D().OnMemoryWrite();
        }
@@ -44,7 +43,7 @@ void KeplerMemory::CallMethod(const GPU::MethodCall& method_call) {
 void KeplerMemory::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
                                   u32 methods_pending) {
    for (std::size_t i = 0; i < amount; i++) {
-        CallMethod({method, base_start[i], 0, methods_pending - static_cast<u32>(i)});
+        CallMethod(method, base_start[i], methods_pending - static_cast<u32>(i) <= 1);
    }
 }
diff --git a/src/video_core/engines/kepler_memory.h b/src/video_core/engines/kepler_memory.h
index bb26fb030..5b7f71a00 100644
--- a/src/video_core/engines/kepler_memory.h
+++ b/src/video_core/engines/kepler_memory.h
@@ -10,6 +10,7 @@
 #include "common/bit_field.h"
 #include "common/common_funcs.h"
 #include "common/common_types.h"
+#include "video_core/engines/engine_interface.h"
 #include "video_core/engines/engine_upload.h"
 #include "video_core/gpu.h"
@@ -32,16 +33,17 @@ namespace Tegra::Engines {
 #define KEPLERMEMORY_REG_INDEX(field_name)                                                         \
    (offsetof(Tegra::Engines::KeplerMemory::Regs, field_name) / sizeof(u32))
-class KeplerMemory final {
+class KeplerMemory final : public EngineInterface {
 public:
    KeplerMemory(Core::System& system, MemoryManager& memory_manager);
    ~KeplerMemory();
    /// Write the value to the register identified by method.
-    void CallMethod(const GPU::MethodCall& method_call);
+    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
    /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending);
+    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+                         u32 methods_pending) override;
    struct Regs {
        static constexpr size_t NUM_REGS = 0x7F;
diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp
index 39e3b66a2..024c9e43b 100644
--- a/src/video_core/engines/maxwell_3d.cpp
+++ b/src/video_core/engines/maxwell_3d.cpp
@@ -44,6 +44,12 @@ void Maxwell3D::InitializeRegisterDefaults() {
        viewport.depth_range_near = 0.0f;
        viewport.depth_range_far = 1.0f;
    }
+    for (auto& viewport : regs.viewport_transform) {
+        viewport.swizzle.x.Assign(Regs::ViewportSwizzle::PositiveX);
+        viewport.swizzle.y.Assign(Regs::ViewportSwizzle::PositiveY);
+        viewport.swizzle.z.Assign(Regs::ViewportSwizzle::PositiveZ);
+        viewport.swizzle.w.Assign(Regs::ViewportSwizzle::PositiveW);
+    }
    // Doom and Bomberman seems to use the uninitialized registers and just enable blend
    // so initialize blend registers with sane values
@@ -125,12 +131,10 @@ void Maxwell3D::CallMacroMethod(u32 method, std::size_t num_parameters, const u3
    }
 }
-void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
+void Maxwell3D::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
-    const u32 method = method_call.method;
    if (method == cb_data_state.current) {
-        regs.reg_array[method] = method_call.argument;
+        regs.reg_array[method] = method_argument;
-        ProcessCBData(method_call.argument);
+        ProcessCBData(method_argument);
        return;
    } else if (cb_data_state.current != null_cb_data) {
        FinishCBData();
@@ -153,10 +157,10 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
            executing_macro = method;
        }
-        macro_params.push_back(method_call.argument);
+        macro_params.push_back(method_argument);
        // Call the macro when there are no more parameters in the command buffer
-        if (method_call.IsLastCall()) {
+        if (is_last_call) {
            CallMacroMethod(executing_macro, macro_params.size(), macro_params.data());
            macro_params.clear();
        }
@@ -166,7 +170,7 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
    ASSERT_MSG(method < Regs::NUM_REGS,
               "Invalid Maxwell3D register, increase the size of the Regs structure");
-    u32 arg = method_call.argument;
+    u32 arg = method_argument;
    // Keep track of the register value in shadow_state when requested.
    if (shadow_state.shadow_ram_control == Regs::ShadowRamControl::Track ||
        shadow_state.shadow_ram_control == Regs::ShadowRamControl::TrackWithFilter) {
@@ -184,8 +188,12 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
    }
    switch (method) {
+    case MAXWELL3D_REG_INDEX(wait_for_idle): {
+        rasterizer.WaitForIdle();
+        break;
+    }
    case MAXWELL3D_REG_INDEX(shadow_ram_control): {
-        shadow_state.shadow_ram_control = static_cast<Regs::ShadowRamControl>(method_call.argument);
+        shadow_state.shadow_ram_control = static_cast<Regs::ShadowRamControl>(method_argument);
        break;
    }
    case MAXWELL3D_REG_INDEX(macros.data): {
@@ -268,7 +276,6 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
        break;
    }
    case MAXWELL3D_REG_INDEX(data_upload): {
-        const bool is_last_call = method_call.IsLastCall();
        upload_state.ProcessData(arg, is_last_call);
        if (is_last_call) {
            OnMemoryWrite();
@@ -326,7 +333,7 @@ void Maxwell3D::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
    }
    default: {
        for (std::size_t i = 0; i < amount; i++) {
-            CallMethod({method, base_start[i], 0, methods_pending - static_cast<u32>(i)});
+            CallMethod(method, base_start[i], methods_pending - static_cast<u32>(i) <= 1);
        }
    }
    }
@@ -356,16 +363,15 @@ void Maxwell3D::StepInstance(const MMEDrawMode expected_mode, const u32 count) {
    StepInstance(expected_mode, count);
 }
-void Maxwell3D::CallMethodFromMME(const GPU::MethodCall& method_call) {
+void Maxwell3D::CallMethodFromMME(u32 method, u32 method_argument) {
-    const u32 method = method_call.method;
    if (mme_inline[method]) {
-        regs.reg_array[method] = method_call.argument;
+        regs.reg_array[method] = method_argument;
        if (method == MAXWELL3D_REG_INDEX(vertex_buffer.count) ||
            method == MAXWELL3D_REG_INDEX(index_array.count)) {
            const MMEDrawMode expected_mode = method == MAXWELL3D_REG_INDEX(vertex_buffer.count)
                                                  ? MMEDrawMode::Array
                                                  : MMEDrawMode::Indexed;
-            StepInstance(expected_mode, method_call.argument);
+            StepInstance(expected_mode, method_argument);
        } else if (method == MAXWELL3D_REG_INDEX(draw.vertex_begin_gl)) {
            mme_draw.instance_mode =
                (regs.draw.instance_next != 0) || (regs.draw.instance_cont != 0);
@@ -377,7 +383,7 @@ void Maxwell3D::CallMethodFromMME(const GPU::MethodCall& method_call) {
        if (mme_draw.current_mode != MMEDrawMode::Undefined) {
            FlushMMEInlineDraw();
        }
-        CallMethod(method_call);
+        CallMethod(method, method_argument, true);
    }
 }
diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h
index 5e522e0d2..05dd6b39b 100644
--- a/src/video_core/engines/maxwell_3d.h
+++ b/src/video_core/engines/maxwell_3d.h
@@ -19,6 +19,7 @@
 #include "common/math_util.h"
 #include "video_core/engines/const_buffer_engine_interface.h"
 #include "video_core/engines/const_buffer_info.h"
+#include "video_core/engines/engine_interface.h"
 #include "video_core/engines/engine_upload.h"
 #include "video_core/engines/shader_type.h"
 #include "video_core/gpu.h"
@@ -48,7 +49,7 @@ namespace Tegra::Engines {
 #define MAXWELL3D_REG_INDEX(field_name)                                                            \
    (offsetof(Tegra::Engines::Maxwell3D::Regs, field_name) / sizeof(u32))
-class Maxwell3D final : public ConstBufferEngineInterface {
+class Maxwell3D final : public ConstBufferEngineInterface, public EngineInterface {
 public:
    explicit Maxwell3D(Core::System& system, VideoCore::RasterizerInterface& rasterizer,
                       MemoryManager& memory_manager);
@@ -575,6 +576,17 @@ public:
            Replay = 3,
        };
+        enum class ViewportSwizzle : u32 {
+            PositiveX = 0,
+            NegativeX = 1,
+            PositiveY = 2,
+            NegativeY = 3,
+            PositiveZ = 4,
+            NegativeZ = 5,
+            PositiveW = 6,
+            NegativeW = 7,
+        };
        struct RenderTargetConfig {
            u32 address_high;
            u32 address_low;
@@ -618,7 +630,14 @@ public:
            f32 translate_x;
            f32 translate_y;
            f32 translate_z;
-            INSERT_UNION_PADDING_WORDS(2);
+            union {
+                u32 raw;
+                BitField<0, 3, ViewportSwizzle> x;
+                BitField<4, 3, ViewportSwizzle> y;
+                BitField<8, 3, ViewportSwizzle> z;
+                BitField<12, 3, ViewportSwizzle> w;
+            } swizzle;
+            INSERT_UNION_PADDING_WORDS(1);
            Common::Rectangle<f32> GetRect() const {
                return {
@@ -709,7 +728,9 @@ public:
        union {
            struct {
-                INSERT_UNION_PADDING_WORDS(0x45);
+                INSERT_UNION_PADDING_WORDS(0x44);
+                u32 wait_for_idle;
                struct {
                    u32 upload_address;
@@ -1358,13 +1379,14 @@ public:
    u32 GetRegisterValue(u32 method) const;
    /// Write the value to the register identified by method.
-    void CallMethod(const GPU::MethodCall& method_call);
+    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
    /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending);
+    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+                         u32 methods_pending) override;
    /// Write the value to the register identified by method.
-    void CallMethodFromMME(const GPU::MethodCall& method_call);
+    void CallMethodFromMME(u32 method, u32 method_argument);
    void FlushMMEInlineDraw();
@@ -1536,6 +1558,7 @@ private:
    static_assert(offsetof(Maxwell3D::Regs, field_name) == position * 4,                           \
                  "Field " #field_name " has invalid position")
+ASSERT_REG_POSITION(wait_for_idle, 0x44);
 ASSERT_REG_POSITION(macros, 0x45);
 ASSERT_REG_POSITION(shadow_ram_control, 0x49);
 ASSERT_REG_POSITION(upload, 0x60);
diff --git a/src/video_core/engines/maxwell_dma.cpp b/src/video_core/engines/maxwell_dma.cpp
index 6630005b0..01d7df405 100644
--- a/src/video_core/engines/maxwell_dma.cpp
+++ b/src/video_core/engines/maxwell_dma.cpp
@@ -17,16 +17,16 @@ namespace Tegra::Engines {
 MaxwellDMA::MaxwellDMA(Core::System& system, MemoryManager& memory_manager)
    : system{system}, memory_manager{memory_manager} {}
-void MaxwellDMA::CallMethod(const GPU::MethodCall& method_call) {
+void MaxwellDMA::CallMethod(u32 method, u32 method_argument, bool is_last_call) {
-    ASSERT_MSG(method_call.method < Regs::NUM_REGS,
+    ASSERT_MSG(method < Regs::NUM_REGS,
               "Invalid MaxwellDMA register, increase the size of the Regs structure");
-    regs.reg_array[method_call.method] = method_call.argument;
+    regs.reg_array[method] = method_argument;
 #define MAXWELLDMA_REG_INDEX(field_name)                                                           \
    (offsetof(Tegra::Engines::MaxwellDMA::Regs, field_name) / sizeof(u32))
-    switch (method_call.method) {
+    switch (method) {
    case MAXWELLDMA_REG_INDEX(exec): {
        HandleCopy();
        break;
@@ -39,7 +39,7 @@ void MaxwellDMA::CallMethod(const GPU::MethodCall& method_call) {
 void MaxwellDMA::CallMultiMethod(u32 method, const u32* base_start, u32 amount,
                                 u32 methods_pending) {
    for (std::size_t i = 0; i < amount; i++) {
-        CallMethod({method, base_start[i], 0, methods_pending - static_cast<u32>(i)});
+        CallMethod(method, base_start[i], methods_pending - static_cast<u32>(i) <= 1);
    }
 }
@@ -90,7 +90,47 @@ void MaxwellDMA::HandleCopy() {
    ASSERT(regs.exec.enable_2d == 1);
    if (regs.exec.is_dst_linear && !regs.exec.is_src_linear) {
        ASSERT(regs.src_params.BlockDepth() == 0);
+        // Optimized path for micro copies.
+        if (regs.dst_pitch * regs.y_count < Texture::GetGOBSize() && regs.dst_pitch <= 64) {
+            const u32 bytes_per_pixel = regs.dst_pitch / regs.x_count;
+            const std::size_t src_size = Texture::GetGOBSize();
+            const std::size_t dst_size = regs.dst_pitch * regs.y_count;
+            u32 pos_x = regs.src_params.pos_x;
+            u32 pos_y = regs.src_params.pos_y;
+            const u64 offset =
+                Texture::GetGOBOffset(regs.src_params.size_x, regs.src_params.size_y, pos_x, pos_y,
+                                      regs.src_params.BlockDepth(), bytes_per_pixel);
+            const u32 x_in_gob = 64 / bytes_per_pixel;
+            pos_x = pos_x % x_in_gob;
+            pos_y = pos_y % 8;
+            if (read_buffer.size() < src_size) {
+                read_buffer.resize(src_size);
+            }
+            if (write_buffer.size() < dst_size) {
+                write_buffer.resize(dst_size);
+            }
+            if (Settings::IsGPULevelExtreme()) {
+                memory_manager.ReadBlock(source + offset, read_buffer.data(), src_size);
+                memory_manager.ReadBlock(dest, write_buffer.data(), dst_size);
+            } else {
+                memory_manager.ReadBlockUnsafe(source + offset, read_buffer.data(), src_size);
+                memory_manager.ReadBlockUnsafe(dest, write_buffer.data(), dst_size);
+            }
+            Texture::UnswizzleSubrect(regs.x_count, regs.y_count, regs.dst_pitch,
+                                      regs.src_params.size_x, bytes_per_pixel, read_buffer.data(),
+                                      write_buffer.data(), regs.src_params.BlockHeight(), pos_x,
+                                      pos_y);
+            memory_manager.WriteBlock(dest, write_buffer.data(), dst_size);
+            return;
+        }
        // If the input is tiled and the output is linear, deswizzle the input and copy it over.
        const u32 bytes_per_pixel = regs.dst_pitch / regs.x_count;
        const std::size_t src_size = Texture::CalculateSize(
diff --git a/src/video_core/engines/maxwell_dma.h b/src/video_core/engines/maxwell_dma.h
index c43ed8194..502dd8509 100644
--- a/src/video_core/engines/maxwell_dma.h
+++ b/src/video_core/engines/maxwell_dma.h
@@ -10,6 +10,7 @@
 #include "common/bit_field.h"
 #include "common/common_funcs.h"
 #include "common/common_types.h"
+#include "video_core/engines/engine_interface.h"
 #include "video_core/gpu.h"
 namespace Core {
@@ -27,16 +28,17 @@ namespace Tegra::Engines {
 * https://github.com/envytools/envytools/blob/master/rnndb/fifo/gk104_copy.xml
 */
-class MaxwellDMA final {
+class MaxwellDMA final : public EngineInterface {
 public:
    explicit MaxwellDMA(Core::System& system, MemoryManager& memory_manager);
    ~MaxwellDMA() = default;
    /// Write the value to the register identified by method.
-    void CallMethod(const GPU::MethodCall& method_call);
+    void CallMethod(u32 method, u32 method_argument, bool is_last_call) override;
    /// Write multiple values to the register identified by method.
-    void CallMultiMethod(u32 method, const u32* base_start, u32 amount, u32 methods_pending);
+    void CallMultiMethod(u32 method, const u32* base_start, u32 amount,
+                         u32 methods_pending) override;
    struct Regs {
        static constexpr std::size_t NUM_REGS = 0x1D6;
diff --git a/src/video_core/gpu.cpp b/src/video_core/gpu.cpp
index b87fd873d..8eb017f65 100644
--- a/src/video_core/gpu.cpp
+++ b/src/video_core/gpu.cpp
@@ -299,19 +299,21 @@ void GPU::CallEngineMethod(const MethodCall& method_call) {
    switch (engine) {
    case EngineID::FERMI_TWOD_A:
-        fermi_2d->CallMethod(method_call);
+        fermi_2d->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
        break;
    case EngineID::MAXWELL_B:
-        maxwell_3d->CallMethod(method_call);
+        maxwell_3d->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
        break;
    case EngineID::KEPLER_COMPUTE_B:
-        kepler_compute->CallMethod(method_call);
+        kepler_compute->CallMethod(method_call.method, method_call.argument,
+                                   method_call.IsLastCall());
        break;
    case EngineID::MAXWELL_DMA_COPY_A:
-        maxwell_dma->CallMethod(method_call);
+        maxwell_dma->CallMethod(method_call.method, method_call.argument, method_call.IsLastCall());
        break;
    case EngineID::KEPLER_INLINE_TO_MEMORY_B:
-        kepler_memory->CallMethod(method_call);
+        kepler_memory->CallMethod(method_call.method, method_call.argument,
+                                  method_call.IsLastCall());
        break;
    default:
        UNIMPLEMENTED_MSG("Unimplemented engine");
@@ -347,7 +349,27 @@ void GPU::ProcessBindMethod(const MethodCall& method_call) {
    // Bind the current subchannel to the desired engine id.
    LOG_DEBUG(HW_GPU, "Binding subchannel {} to engine {}", method_call.subchannel,
              method_call.argument);
-    bound_engines[method_call.subchannel] = static_cast<EngineID>(method_call.argument);
+    const auto engine_id = static_cast<EngineID>(method_call.argument);
+    bound_engines[method_call.subchannel] = static_cast<EngineID>(engine_id);
+    switch (engine_id) {
+    case EngineID::FERMI_TWOD_A:
+        dma_pusher->BindSubchannel(fermi_2d.get(), method_call.subchannel);
+        break;
+    case EngineID::MAXWELL_B:
+        dma_pusher->BindSubchannel(maxwell_3d.get(), method_call.subchannel);
+        break;
+    case EngineID::KEPLER_COMPUTE_B:
+        dma_pusher->BindSubchannel(kepler_compute.get(), method_call.subchannel);
+        break;
+    case EngineID::MAXWELL_DMA_COPY_A:
+        dma_pusher->BindSubchannel(maxwell_dma.get(), method_call.subchannel);
+        break;
+    case EngineID::KEPLER_INLINE_TO_MEMORY_B:
+        dma_pusher->BindSubchannel(kepler_memory.get(), method_call.subchannel);
+        break;
+    default:
+        UNIMPLEMENTED_MSG("Unimplemented engine {:04X}", static_cast<u32>(engine_id));
+    }
 }
 void GPU::ProcessSemaphoreTriggerMethod() {
diff --git a/src/video_core/gpu.h b/src/video_core/gpu.h
index dd51c95b7..a1b4c305c 100644
--- a/src/video_core/gpu.h
+++ b/src/video_core/gpu.h
@@ -68,6 +68,7 @@ enum class RenderTargetFormat : u32 {
    BGR5A1_UNORM = 0xE9,
    RG8_UNORM = 0xEA,
    RG8_SNORM = 0xEB,
+    RG8_UINT = 0xED,
    R16_UNORM = 0xEE,
    R16_SNORM = 0xEF,
    R16_SINT = 0xF0,
diff --git a/src/video_core/macro_interpreter.cpp b/src/video_core/macro_interpreter.cpp
index 42031d80a..947364928 100644
--- a/src/video_core/macro_interpreter.cpp
+++ b/src/video_core/macro_interpreter.cpp
@@ -328,7 +328,7 @@ void MacroInterpreter::SetMethodAddress(u32 address) {
 }
 void MacroInterpreter::Send(u32 value) {
-    maxwell3d.CallMethodFromMME({method_address.address, value});
+    maxwell3d.CallMethodFromMME(method_address.address, value);
    // Increment the method address by the method increment.
    method_address.address.Assign(method_address.address.Value() +
                                  method_address.increment.Value());
diff --git a/src/video_core/morton.cpp b/src/video_core/morton.cpp
index 6d522c318..836b25c1d 100644
--- a/src/video_core/morton.cpp
+++ b/src/video_core/morton.cpp
@@ -83,6 +83,7 @@ static constexpr ConversionArray morton_to_linear_fns = {
    MortonCopy<true, PixelFormat::RGBA8_SRGB>,
    MortonCopy<true, PixelFormat::RG8U>,
    MortonCopy<true, PixelFormat::RG8S>,
+    MortonCopy<true, PixelFormat::RG8UI>,
    MortonCopy<true, PixelFormat::RG32UI>,
    MortonCopy<true, PixelFormat::RGBX16F>,
    MortonCopy<true, PixelFormat::R32UI>,
@@ -166,6 +167,7 @@ static constexpr ConversionArray linear_to_morton_fns = {
    MortonCopy<false, PixelFormat::RGBA8_SRGB>,
    MortonCopy<false, PixelFormat::RG8U>,
    MortonCopy<false, PixelFormat::RG8S>,
+    MortonCopy<false, PixelFormat::RG8UI>,
    MortonCopy<false, PixelFormat::RG32UI>,
    MortonCopy<false, PixelFormat::RGBX16F>,
    MortonCopy<false, PixelFormat::R32UI>,
diff --git a/src/video_core/rasterizer_interface.h b/src/video_core/rasterizer_interface.h
index 603f61952..3cbdac8e7 100644
--- a/src/video_core/rasterizer_interface.h
+++ b/src/video_core/rasterizer_interface.h
@@ -80,6 +80,9 @@ public:
    /// and invalidated
    virtual void FlushAndInvalidateRegion(VAddr addr, u64 size) = 0;
+    /// Notify the host renderer to wait for previous primitive and compute operations.
+    virtual void WaitForIdle() = 0;
    /// Notify the rasterizer to send all written commands to the host GPU.
    virtual void FlushCommands() = 0;
diff --git a/src/video_core/renderer_opengl/gl_buffer_cache.cpp b/src/video_core/renderer_opengl/gl_buffer_cache.cpp
index 4efce0de7..d2cab50bd 100644
--- a/src/video_core/renderer_opengl/gl_buffer_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_buffer_cache.cpp
@@ -51,10 +51,6 @@ Buffer OGLBufferCache::CreateBlock(VAddr cpu_addr, std::size_t size) {
    return std::make_shared<CachedBufferBlock>(cpu_addr, size);
 }
-void OGLBufferCache::WriteBarrier() {
-    glMemoryBarrier(GL_SHADER_STORAGE_BARRIER_BIT);
-}
 GLuint OGLBufferCache::ToHandle(const Buffer& buffer) {
    return buffer->GetHandle();
 }
diff --git a/src/video_core/renderer_opengl/gl_buffer_cache.h b/src/video_core/renderer_opengl/gl_buffer_cache.h
index a74817857..a9e86cfc7 100644
--- a/src/video_core/renderer_opengl/gl_buffer_cache.h
+++ b/src/video_core/renderer_opengl/gl_buffer_cache.h
@@ -59,8 +59,6 @@ protected:
    GLuint ToHandle(const Buffer& buffer) override;
-    void WriteBarrier() override;
    void UploadBlockData(const Buffer& buffer, std::size_t offset, std::size_t size,
                         const u8* data) override;
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp
index 725b4c32d..69dcf952f 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp
@@ -746,6 +746,17 @@ void RasterizerOpenGL::FlushAndInvalidateRegion(VAddr addr, u64 size) {
    InvalidateRegion(addr, size);
 }
+void RasterizerOpenGL::WaitForIdle() {
+    // Place a barrier on everything that is not framebuffer related.
+    // This is related to another flag that is not currently implemented.
+    glMemoryBarrier(GL_VERTEX_ATTRIB_ARRAY_BARRIER_BIT | GL_ELEMENT_ARRAY_BARRIER_BIT |
+                    GL_UNIFORM_BARRIER_BIT | GL_TEXTURE_FETCH_BARRIER_BIT |
+                    GL_SHADER_IMAGE_ACCESS_BARRIER_BIT | GL_COMMAND_BARRIER_BIT |
+                    GL_PIXEL_BUFFER_BARRIER_BIT | GL_TEXTURE_UPDATE_BARRIER_BIT |
+                    GL_BUFFER_UPDATE_BARRIER_BIT | GL_TRANSFORM_FEEDBACK_BARRIER_BIT |
+                    GL_SHADER_STORAGE_BARRIER_BIT | GL_QUERY_BUFFER_BARRIER_BIT);
+}
 void RasterizerOpenGL::FlushCommands() {
    // Only flush when we have commands queued to OpenGL.
    if (num_queued_commands == 0) {
@@ -1008,6 +1019,14 @@ void RasterizerOpenGL::SyncViewport() {
            const GLdouble near_depth = src.translate_z - src.scale_z * reduce_z;
            const GLdouble far_depth = src.translate_z + src.scale_z;
            glDepthRangeIndexed(static_cast<GLuint>(i), near_depth, far_depth);
+            if (!GLAD_GL_NV_viewport_swizzle) {
+                continue;
+            }
+            glViewportSwizzleNV(static_cast<GLuint>(i), MaxwellToGL::ViewportSwizzle(src.swizzle.x),
+                                MaxwellToGL::ViewportSwizzle(src.swizzle.y),
+                                MaxwellToGL::ViewportSwizzle(src.swizzle.z),
+                                MaxwellToGL::ViewportSwizzle(src.swizzle.w));
        }
    }
 }
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.h b/src/video_core/renderer_opengl/gl_rasterizer.h
index 87249fb6f..b94c65907 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.h
+++ b/src/video_core/renderer_opengl/gl_rasterizer.h
@@ -75,6 +75,7 @@ public:
    void SignalSyncPoint(u32 value) override;
    void ReleaseFences() override;
    void FlushAndInvalidateRegion(VAddr addr, u64 size) override;
+    void WaitForIdle() override;
    void FlushCommands() override;
    void TickFrame() override;
    bool AccelerateSurfaceCopy(const Tegra::Engines::Fermi2D::Regs::Surface& src,
diff --git a/src/video_core/renderer_opengl/gl_texture_cache.cpp b/src/video_core/renderer_opengl/gl_texture_cache.cpp
index 2729d1265..94fbd2a22 100644
--- a/src/video_core/renderer_opengl/gl_texture_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_texture_cache.cpp
@@ -83,6 +83,7 @@ constexpr std::array<FormatTuple, VideoCore::Surface::MaxPixelFormat> tex_format
    {GL_SRGB8_ALPHA8, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8_REV},      // RGBA8_SRGB
    {GL_RG8, GL_RG, GL_UNSIGNED_BYTE},                            // RG8U
    {GL_RG8_SNORM, GL_RG, GL_BYTE},                               // RG8S
+    {GL_RG8UI, GL_RG_INTEGER, GL_UNSIGNED_INT},                   // RG8UI
    {GL_RG32UI, GL_RG_INTEGER, GL_UNSIGNED_INT},                  // RG32UI
    {GL_RGB16F, GL_RGBA, GL_HALF_FLOAT},                          // RGBX16F
    {GL_R32UI, GL_RED_INTEGER, GL_UNSIGNED_INT},                  // R32UI
diff --git a/src/video_core/renderer_opengl/maxwell_to_gl.h b/src/video_core/renderer_opengl/maxwell_to_gl.h
index 2c0c77c28..994ae98eb 100644
--- a/src/video_core/renderer_opengl/maxwell_to_gl.h
+++ b/src/video_core/renderer_opengl/maxwell_to_gl.h
@@ -503,5 +503,10 @@ inline GLenum PolygonMode(Maxwell::PolygonMode polygon_mode) {
    return GL_FILL;
 }
+inline GLenum ViewportSwizzle(Maxwell::ViewportSwizzle swizzle) {
+    // Enumeration order matches register order. We can convert it arithmetically.
+    return GL_VIEWPORT_SWIZZLE_POSITIVE_X_NV + static_cast<GLenum>(swizzle);
+}
 } // namespace MaxwellToGL
 } // namespace OpenGL
diff --git a/src/video_core/renderer_vulkan/fixed_pipeline_state.cpp b/src/video_core/renderer_vulkan/fixed_pipeline_state.cpp
index 7b01adf7f..568744e3c 100644
--- a/src/video_core/renderer_vulkan/fixed_pipeline_state.cpp
+++ b/src/video_core/renderer_vulkan/fixed_pipeline_state.cpp
@@ -2,6 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
+#include <algorithm>
 #include <cstring>
 #include <tuple>
@@ -102,6 +103,12 @@ void FixedPipelineState::ColorBlending::Fill(const Maxwell& regs) noexcept {
    }
 }
+void FixedPipelineState::ViewportSwizzles::Fill(const Maxwell& regs) noexcept {
+    const auto& transform = regs.viewport_transform;
+    std::transform(transform.begin(), transform.end(), swizzles.begin(),
+                   [](const auto& viewport) { return static_cast<u16>(viewport.swizzle.raw); });
+}
 void FixedPipelineState::BlendingAttachment::Fill(const Maxwell& regs, std::size_t index) {
    const auto& mask = regs.color_mask[regs.color_mask_common ? 0 : index];
@@ -145,6 +152,7 @@ void FixedPipelineState::Fill(const Maxwell& regs) {
    rasterizer.Fill(regs);
    depth_stencil.Fill(regs);
    color_blending.Fill(regs);
+    viewport_swizzles.Fill(regs);
 }
 std::size_t FixedPipelineState::Hash() const noexcept {
diff --git a/src/video_core/renderer_vulkan/fixed_pipeline_state.h b/src/video_core/renderer_vulkan/fixed_pipeline_state.h
index cbf55dda3..31a6398f2 100644
--- a/src/video_core/renderer_vulkan/fixed_pipeline_state.h
+++ b/src/video_core/renderer_vulkan/fixed_pipeline_state.h
@@ -234,10 +234,17 @@ struct FixedPipelineState {
        void Fill(const Maxwell& regs) noexcept;
    };
+    struct ViewportSwizzles {
+        std::array<u16, Maxwell::NumViewports> swizzles;
+        void Fill(const Maxwell& regs) noexcept;
+    };
    VertexInput vertex_input;
    Rasterizer rasterizer;
    DepthStencil depth_stencil;
    ColorBlending color_blending;
+    ViewportSwizzles viewport_swizzles;
    void Fill(const Maxwell& regs);
diff --git a/src/video_core/renderer_vulkan/maxwell_to_vk.cpp b/src/video_core/renderer_vulkan/maxwell_to_vk.cpp
index 8681b821f..12be691a5 100644
--- a/src/video_core/renderer_vulkan/maxwell_to_vk.cpp
+++ b/src/video_core/renderer_vulkan/maxwell_to_vk.cpp
@@ -160,6 +160,7 @@ struct FormatTuple {
    {VK_FORMAT_R8G8B8A8_SRGB, Attachable},                      // RGBA8_SRGB
    {VK_FORMAT_R8G8_UNORM, Attachable | Storage},               // RG8U
    {VK_FORMAT_R8G8_SNORM, Attachable | Storage},               // RG8S
+    {VK_FORMAT_R8G8_UINT, Attachable | Storage},                // RG8UI
    {VK_FORMAT_R32G32_UINT, Attachable | Storage},              // RG32UI
    {VK_FORMAT_UNDEFINED},                                      // RGBX16F
    {VK_FORMAT_R32_UINT, Attachable | Storage},                 // R32UI
@@ -345,8 +346,6 @@ VkFormat VertexFormat(Maxwell::VertexAttribute::Type type, Maxwell::VertexAttrib
        break;
    case Maxwell::VertexAttribute::Type::SignedInt:
        switch (size) {
-        case Maxwell::VertexAttribute::Size::Size_16_16_16_16:
-            return VK_FORMAT_R16G16B16A16_SINT;
        case Maxwell::VertexAttribute::Size::Size_8:
            return VK_FORMAT_R8_SINT;
        case Maxwell::VertexAttribute::Size::Size_8_8:
@@ -355,8 +354,22 @@ VkFormat VertexFormat(Maxwell::VertexAttribute::Type type, Maxwell::VertexAttrib
            return VK_FORMAT_R8G8B8_SINT;
        case Maxwell::VertexAttribute::Size::Size_8_8_8_8:
            return VK_FORMAT_R8G8B8A8_SINT;
+        case Maxwell::VertexAttribute::Size::Size_16:
+            return VK_FORMAT_R16_SINT;
+        case Maxwell::VertexAttribute::Size::Size_16_16:
+            return VK_FORMAT_R16G16_SINT;
+        case Maxwell::VertexAttribute::Size::Size_16_16_16:
+            return VK_FORMAT_R16G16B16_SINT;
+        case Maxwell::VertexAttribute::Size::Size_16_16_16_16:
+            return VK_FORMAT_R16G16B16A16_SINT;
        case Maxwell::VertexAttribute::Size::Size_32:
            return VK_FORMAT_R32_SINT;
+        case Maxwell::VertexAttribute::Size::Size_32_32:
+            return VK_FORMAT_R32G32_SINT;
+        case Maxwell::VertexAttribute::Size::Size_32_32_32:
+            return VK_FORMAT_R32G32B32_SINT;
+        case Maxwell::VertexAttribute::Size::Size_32_32_32_32:
+            return VK_FORMAT_R32G32B32A32_SINT;
        default:
            break;
        }
@@ -672,4 +685,27 @@ VkComponentSwizzle SwizzleSource(Tegra::Texture::SwizzleSource swizzle) {
    return {};
 }
+VkViewportCoordinateSwizzleNV ViewportSwizzle(Maxwell::ViewportSwizzle swizzle) {
+    switch (swizzle) {
+    case Maxwell::ViewportSwizzle::PositiveX:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_X_NV;
+    case Maxwell::ViewportSwizzle::NegativeX:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_X_NV;
+    case Maxwell::ViewportSwizzle::PositiveY:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Y_NV;
+    case Maxwell::ViewportSwizzle::NegativeY:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_Y_NV;
+    case Maxwell::ViewportSwizzle::PositiveZ:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_Z_NV;
+    case Maxwell::ViewportSwizzle::NegativeZ:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_Z_NV;
+    case Maxwell::ViewportSwizzle::PositiveW:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_POSITIVE_W_NV;
+    case Maxwell::ViewportSwizzle::NegativeW:
+        return VK_VIEWPORT_COORDINATE_SWIZZLE_NEGATIVE_W_NV;
+    }
+    UNREACHABLE_MSG("Invalid swizzle={}", static_cast<int>(swizzle));
+    return {};
+}
 } // namespace Vulkan::MaxwellToVK
diff --git a/src/video_core/renderer_vulkan/maxwell_to_vk.h b/src/video_core/renderer_vulkan/maxwell_to_vk.h
index 81bce4c6c..7e213452f 100644
--- a/src/video_core/renderer_vulkan/maxwell_to_vk.h
+++ b/src/video_core/renderer_vulkan/maxwell_to_vk.h
@@ -59,4 +59,6 @@ VkCullModeFlags CullFace(Maxwell::CullFace cull_face);
 VkComponentSwizzle SwizzleSource(Tegra::Texture::SwizzleSource swizzle);
+VkViewportCoordinateSwizzleNV ViewportSwizzle(Maxwell::ViewportSwizzle swizzle);
 } // namespace Vulkan::MaxwellToVK
diff --git a/src/video_core/renderer_vulkan/renderer_vulkan.cpp b/src/video_core/renderer_vulkan/renderer_vulkan.cpp
index 04532f8f8..59b441943 100644
--- a/src/video_core/renderer_vulkan/renderer_vulkan.cpp
+++ b/src/video_core/renderer_vulkan/renderer_vulkan.cpp
@@ -12,15 +12,12 @@
 #include <fmt/format.h>
-#include "common/assert.h"
 #include "common/dynamic_library.h"
 #include "common/logging/log.h"
 #include "common/telemetry.h"
 #include "core/core.h"
 #include "core/core_timing.h"
 #include "core/frontend/emu_window.h"
-#include "core/memory.h"
-#include "core/perf_stats.h"
 #include "core/settings.h"
 #include "core/telemetry_session.h"
 #include "video_core/gpu.h"
diff --git a/src/video_core/renderer_vulkan/renderer_vulkan.h b/src/video_core/renderer_vulkan/renderer_vulkan.h
index 18270909b..522b5bff8 100644
--- a/src/video_core/renderer_vulkan/renderer_vulkan.h
+++ b/src/video_core/renderer_vulkan/renderer_vulkan.h
@@ -5,7 +5,6 @@
 #pragma once
 #include <memory>
-#include <optional>
 #include <string>
 #include <vector>
diff --git a/src/video_core/renderer_vulkan/vk_blit_screen.h b/src/video_core/renderer_vulkan/vk_blit_screen.h
index 5eb544aea..243640fab 100644
--- a/src/video_core/renderer_vulkan/vk_blit_screen.h
+++ b/src/video_core/renderer_vulkan/vk_blit_screen.h
@@ -4,7 +4,6 @@
 #pragma once
-#include <array>
 #include <memory>
 #include <tuple>
diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp
index 81e1de2be..5b494da8c 100644
--- a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp
@@ -5,11 +5,7 @@
 #include <algorithm>
 #include <cstring>
 #include <memory>
-#include <optional>
-#include <tuple>
-#include "common/assert.h"
-#include "common/bit_util.h"
 #include "core/core.h"
 #include "video_core/renderer_vulkan/vk_buffer_cache.h"
 #include "video_core/renderer_vulkan/vk_device.h"
diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.h b/src/video_core/renderer_vulkan/vk_buffer_cache.h
index 3cd2e2774..a54583e7d 100644
--- a/src/video_core/renderer_vulkan/vk_buffer_cache.h
+++ b/src/video_core/renderer_vulkan/vk_buffer_cache.h
@@ -5,14 +5,11 @@
 #pragma once
 #include <memory>
-#include <unordered_map>
-#include <vector>
 #include "common/common_types.h"
 #include "video_core/buffer_cache/buffer_cache.h"
 #include "video_core/rasterizer_cache.h"
 #include "video_core/renderer_vulkan/vk_memory_manager.h"
-#include "video_core/renderer_vulkan/vk_resource_manager.h"
 #include "video_core/renderer_vulkan/vk_staging_buffer_pool.h"
 #include "video_core/renderer_vulkan/vk_stream_buffer.h"
 #include "video_core/renderer_vulkan/wrapper.h"
@@ -55,8 +52,6 @@ public:
 protected:
    VkBuffer ToHandle(const Buffer& buffer) override;
-    void WriteBarrier() override {}
    Buffer CreateBlock(VAddr cpu_addr, std::size_t size) override;
    void UploadBlockData(const Buffer& buffer, std::size_t offset, std::size_t size,
diff --git a/src/video_core/renderer_vulkan/vk_compute_pass.cpp b/src/video_core/renderer_vulkan/vk_compute_pass.cpp
index 7b0268033..da71e710c 100644
--- a/src/video_core/renderer_vulkan/vk_compute_pass.cpp
+++ b/src/video_core/renderer_vulkan/vk_compute_pass.cpp
@@ -6,7 +6,7 @@
 #include <memory>
 #include <optional>
 #include <utility>
-#include <vector>
 #include "common/alignment.h"
 #include "common/assert.h"
 #include "common/common_types.h"
diff --git a/src/video_core/renderer_vulkan/vk_compute_pass.h b/src/video_core/renderer_vulkan/vk_compute_pass.h
index 26bf834de..230b526bc 100644
--- a/src/video_core/renderer_vulkan/vk_compute_pass.h
+++ b/src/video_core/renderer_vulkan/vk_compute_pass.h
@@ -6,7 +6,7 @@
 #include <optional>
 #include <utility>
-#include <vector>
 #include "common/common_types.h"
 #include "video_core/engines/maxwell_3d.h"
 #include "video_core/renderer_vulkan/vk_descriptor_pool.h"
diff --git a/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp b/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp
index 52566bb79..8e1b46277 100644
--- a/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp
+++ b/src/video_core/renderer_vulkan/vk_compute_pipeline.cpp
@@ -2,14 +2,12 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <memory>
 #include <vector>
 #include "video_core/renderer_vulkan/vk_compute_pipeline.h"
 #include "video_core/renderer_vulkan/vk_descriptor_pool.h"
 #include "video_core/renderer_vulkan/vk_device.h"
 #include "video_core/renderer_vulkan/vk_pipeline_cache.h"
-#include "video_core/renderer_vulkan/vk_resource_manager.h"
 #include "video_core/renderer_vulkan/vk_scheduler.h"
 #include "video_core/renderer_vulkan/vk_shader_decompiler.h"
 #include "video_core/renderer_vulkan/vk_update_descriptor.h"
diff --git a/src/video_core/renderer_vulkan/vk_compute_pipeline.h b/src/video_core/renderer_vulkan/vk_compute_pipeline.h
index 33b9af29e..6e2f22a4a 100644
--- a/src/video_core/renderer_vulkan/vk_compute_pipeline.h
+++ b/src/video_core/renderer_vulkan/vk_compute_pipeline.h
@@ -4,8 +4,6 @@
 #pragma once
-#include <memory>
 #include "common/common_types.h"
 #include "video_core/renderer_vulkan/vk_descriptor_pool.h"
 #include "video_core/renderer_vulkan/vk_shader_decompiler.h"
diff --git a/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp b/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp
index e9d528aa6..890fd52cf 100644
--- a/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp
+++ b/src/video_core/renderer_vulkan/vk_descriptor_pool.cpp
@@ -2,7 +2,6 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <memory>
 #include <vector>
 #include "common/common_types.h"
diff --git a/src/video_core/renderer_vulkan/vk_descriptor_pool.h b/src/video_core/renderer_vulkan/vk_descriptor_pool.h
index ab40c70f0..9efa66bef 100644
--- a/src/video_core/renderer_vulkan/vk_descriptor_pool.h
+++ b/src/video_core/renderer_vulkan/vk_descriptor_pool.h
@@ -4,10 +4,8 @@
 #pragma once
-#include <memory>
 #include <vector>
-#include "common/common_types.h"
 #include "video_core/renderer_vulkan/vk_resource_manager.h"
 #include "video_core/renderer_vulkan/wrapper.h"
diff --git a/src/video_core/renderer_vulkan/vk_device.cpp b/src/video_core/renderer_vulkan/vk_device.cpp
index e90c76492..f0c491d00 100644
--- a/src/video_core/renderer_vulkan/vk_device.cpp
+++ b/src/video_core/renderer_vulkan/vk_device.cpp
@@ -4,7 +4,6 @@
 #include <bitset>
 #include <chrono>
-#include <cstdlib>
 #include <optional>
 #include <string_view>
 #include <thread>
@@ -95,6 +94,7 @@ std::unordered_map<VkFormat, VkFormatProperties> GetFormatProperties(
                                        VK_FORMAT_R8G8B8A8_SRGB,
                                        VK_FORMAT_R8G8_UNORM,
                                        VK_FORMAT_R8G8_SNORM,
+                                        VK_FORMAT_R8G8_UINT,
                                        VK_FORMAT_R8_UNORM,
                                        VK_FORMAT_R8_UINT,
                                        VK_FORMAT_B10G11R11_UFLOAT_PACK32,
@@ -261,6 +261,10 @@ bool VKDevice::Create() {
        LOG_INFO(Render_Vulkan, "Device doesn't support float16 natively");
    }
+    if (!nv_viewport_swizzle) {
+        LOG_INFO(Render_Vulkan, "Device doesn't support viewport swizzles");
+    }
    VkPhysicalDeviceUniformBufferStandardLayoutFeaturesKHR std430_layout;
    if (khr_uniform_buffer_standard_layout) {
        std430_layout.sType =
@@ -294,6 +298,17 @@ bool VKDevice::Create() {
        LOG_INFO(Render_Vulkan, "Device doesn't support transform feedbacks");
    }
+    VkPhysicalDeviceCustomBorderColorFeaturesEXT custom_border;
+    if (ext_custom_border_color) {
+        custom_border.sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_CUSTOM_BORDER_COLOR_FEATURES_EXT;
+        custom_border.pNext = nullptr;
+        custom_border.customBorderColors = VK_TRUE;
+        custom_border.customBorderColorWithoutFormat = VK_TRUE;
+        SetNext(next, custom_border);
+    } else {
+        LOG_INFO(Render_Vulkan, "Device doesn't support custom border colors");
+    }
    if (!ext_depth_range_unrestricted) {
        LOG_INFO(Render_Vulkan, "Device doesn't support depth range unrestricted");
    }
@@ -521,7 +536,9 @@ std::vector<const char*> VKDevice::LoadExtensions() {
    bool has_khr_shader_float16_int8{};
    bool has_ext_subgroup_size_control{};
    bool has_ext_transform_feedback{};
+    bool has_ext_custom_border_color{};
    for (const auto& extension : physical.EnumerateDeviceExtensionProperties()) {
+        Test(extension, nv_viewport_swizzle, VK_NV_VIEWPORT_SWIZZLE_EXTENSION_NAME, true);
        Test(extension, khr_uniform_buffer_standard_layout,
             VK_KHR_UNIFORM_BUFFER_STANDARD_LAYOUT_EXTENSION_NAME, true);
        Test(extension, has_khr_shader_float16_int8, VK_KHR_SHADER_FLOAT16_INT8_EXTENSION_NAME,
@@ -535,6 +552,8 @@ std::vector<const char*> VKDevice::LoadExtensions() {
             false);
        Test(extension, has_ext_transform_feedback, VK_EXT_TRANSFORM_FEEDBACK_EXTENSION_NAME,
             false);
+        Test(extension, has_ext_custom_border_color, VK_EXT_CUSTOM_BORDER_COLOR_EXTENSION_NAME,
+             false);
        if (Settings::values.renderer_debug) {
            Test(extension, nv_device_diagnostics_config,
                 VK_NV_DEVICE_DIAGNOSTICS_CONFIG_EXTENSION_NAME, true);
@@ -607,6 +626,19 @@ std::vector<const char*> VKDevice::LoadExtensions() {
        }
    }
+    if (has_ext_custom_border_color) {
+        VkPhysicalDeviceCustomBorderColorFeaturesEXT border_features;
+        border_features.sType = VK_STRUCTURE_TYPE_PHYSICAL_DEVICE_CUSTOM_BORDER_COLOR_FEATURES_EXT;
+        border_features.pNext = nullptr;
+        features.pNext = &border_features;
+        physical.GetFeatures2KHR(features);
+        if (border_features.customBorderColors && border_features.customBorderColorWithoutFormat) {
+            extensions.push_back(VK_EXT_CUSTOM_BORDER_COLOR_EXTENSION_NAME);
+            ext_custom_border_color = true;
+        }
+    }
    return extensions;
 }
diff --git a/src/video_core/renderer_vulkan/vk_device.h b/src/video_core/renderer_vulkan/vk_device.h
index c8640762d..6b9227b09 100644
--- a/src/video_core/renderer_vulkan/vk_device.h
+++ b/src/video_core/renderer_vulkan/vk_device.h
@@ -147,6 +147,11 @@ public:
        return is_formatless_image_load_supported;
    }
+    /// Returns true if the device supports VK_NV_viewport_swizzle.
+    bool IsNvViewportSwizzleSupported() const {
+        return nv_viewport_swizzle;
+    }
    /// Returns true if the device supports VK_EXT_scalar_block_layout.
    bool IsKhrUniformBufferStandardLayoutSupported() const {
        return khr_uniform_buffer_standard_layout;
@@ -172,6 +177,11 @@ public:
        return ext_transform_feedback;
    }
+    /// Returns true if the device supports VK_EXT_custom_border_color.
+    bool IsExtCustomBorderColorSupported() const {
+        return ext_custom_border_color;
+    }
    /// Returns the vendor name reported from Vulkan.
    std::string_view GetVendorName() const {
        return vendor_name;
@@ -222,11 +232,13 @@ private:
    bool is_float16_supported{};            ///< Support for float16 arithmetics.
    bool is_warp_potentially_bigger{};      ///< Host warp size can be bigger than guest.
    bool is_formatless_image_load_supported{}; ///< Support for shader image read without format.
+    bool nv_viewport_swizzle{};                ///< Support for VK_NV_viewport_swizzle.
    bool khr_uniform_buffer_standard_layout{}; ///< Support for std430 on UBOs.
    bool ext_index_type_uint8{};               ///< Support for VK_EXT_index_type_uint8.
    bool ext_depth_range_unrestricted{};       ///< Support for VK_EXT_depth_range_unrestricted.
    bool ext_shader_viewport_index_layer{};    ///< Support for VK_EXT_shader_viewport_index_layer.
    bool ext_transform_feedback{};             ///< Support for VK_EXT_transform_feedback.
+    bool ext_custom_border_color{};            ///< Support for VK_EXT_custom_border_color.
    bool nv_device_diagnostics_config{};       ///< Support for VK_NV_device_diagnostics_config.
    // Telemetry parameters
diff --git a/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp b/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp
index b39b81b48..69b6bba00 100644
--- a/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp
+++ b/src/video_core/renderer_vulkan/vk_graphics_pipeline.cpp
@@ -2,11 +2,11 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
+#include <algorithm>
 #include <array>
 #include <cstring>
 #include <vector>
-#include "common/assert.h"
 #include "common/common_types.h"
 #include "common/microprofile.h"
 #include "video_core/renderer_vulkan/fixed_pipeline_state.h"
@@ -51,6 +51,23 @@ bool SupportsPrimitiveRestart(VkPrimitiveTopology topology) {
                     topology) == std::end(unsupported_topologies);
 }
+VkViewportSwizzleNV UnpackViewportSwizzle(u16 swizzle) {
+    union {
+        u32 raw;
+        BitField<0, 3, Maxwell::ViewportSwizzle> x;
+        BitField<4, 3, Maxwell::ViewportSwizzle> y;
+        BitField<8, 3, Maxwell::ViewportSwizzle> z;
+        BitField<12, 3, Maxwell::ViewportSwizzle> w;
+    } const unpacked{swizzle};
+    VkViewportSwizzleNV result;
+    result.x = MaxwellToVK::ViewportSwizzle(unpacked.x);
+    result.y = MaxwellToVK::ViewportSwizzle(unpacked.y);
+    result.z = MaxwellToVK::ViewportSwizzle(unpacked.z);
+    result.w = MaxwellToVK::ViewportSwizzle(unpacked.w);
+    return result;
+}
 } // Anonymous namespace
 VKGraphicsPipeline::VKGraphicsPipeline(const VKDevice& device, VKScheduler& scheduler,
@@ -163,6 +180,7 @@ vk::Pipeline VKGraphicsPipeline::CreatePipeline(const RenderPassParams& renderpa
    const auto& ds = fixed_state.depth_stencil;
    const auto& cd = fixed_state.color_blending;
    const auto& rs = fixed_state.rasterizer;
+    const auto& viewport_swizzles = fixed_state.viewport_swizzles.swizzles;
    std::vector<VkVertexInputBindingDescription> vertex_bindings;
    std::vector<VkVertexInputBindingDivisorDescriptionEXT> vertex_binding_divisors;
@@ -245,6 +263,19 @@ vk::Pipeline VKGraphicsPipeline::CreatePipeline(const RenderPassParams& renderpa
    viewport_ci.scissorCount = Maxwell::NumViewports;
    viewport_ci.pScissors = nullptr;
+    std::array<VkViewportSwizzleNV, Maxwell::NumViewports> swizzles;
+    std::transform(viewport_swizzles.begin(), viewport_swizzles.end(), swizzles.begin(),
+                   UnpackViewportSwizzle);
+    VkPipelineViewportSwizzleStateCreateInfoNV swizzle_ci;
+    swizzle_ci.sType = VK_STRUCTURE_TYPE_PIPELINE_VIEWPORT_SWIZZLE_STATE_CREATE_INFO_NV;
+    swizzle_ci.pNext = nullptr;
+    swizzle_ci.flags = 0;
+    swizzle_ci.viewportCount = Maxwell::NumViewports;
+    swizzle_ci.pViewportSwizzles = swizzles.data();
+    if (device.IsNvViewportSwizzleSupported()) {
+        viewport_ci.pNext = &swizzle_ci;
+    }
    VkPipelineRasterizationStateCreateInfo rasterization_ci;
    rasterization_ci.sType = VK_STRUCTURE_TYPE_PIPELINE_RASTERIZATION_STATE_CREATE_INFO;
    rasterization_ci.pNext = nullptr;
diff --git a/src/video_core/renderer_vulkan/vk_graphics_pipeline.h b/src/video_core/renderer_vulkan/vk_graphics_pipeline.h
index 7aba70960..a1d699a6c 100644
--- a/src/video_core/renderer_vulkan/vk_graphics_pipeline.h
+++ b/src/video_core/renderer_vulkan/vk_graphics_pipeline.h
@@ -5,16 +5,13 @@
 #pragma once
 #include <array>
-#include <memory>
 #include <optional>
-#include <unordered_map>
 #include <vector>
 #include "video_core/engines/maxwell_3d.h"
 #include "video_core/renderer_vulkan/fixed_pipeline_state.h"
 #include "video_core/renderer_vulkan/vk_descriptor_pool.h"
 #include "video_core/renderer_vulkan/vk_renderpass_cache.h"
-#include "video_core/renderer_vulkan/vk_resource_manager.h"
 #include "video_core/renderer_vulkan/vk_shader_decompiler.h"
 #include "video_core/renderer_vulkan/wrapper.h"
diff --git a/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp b/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp
index 8fbd63dbc..fe45ed269 100644
--- a/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp
@@ -22,7 +22,6 @@
 #include "video_core/renderer_vulkan/vk_pipeline_cache.h"
 #include "video_core/renderer_vulkan/vk_rasterizer.h"
 #include "video_core/renderer_vulkan/vk_renderpass_cache.h"
-#include "video_core/renderer_vulkan/vk_resource_manager.h"
 #include "video_core/renderer_vulkan/vk_scheduler.h"
 #include "video_core/renderer_vulkan/vk_update_descriptor.h"
 #include "video_core/renderer_vulkan/wrapper.h"
diff --git a/src/video_core/renderer_vulkan/vk_pipeline_cache.h b/src/video_core/renderer_vulkan/vk_pipeline_cache.h
index ebddafb73..0b5796fef 100644
--- a/src/video_core/renderer_vulkan/vk_pipeline_cache.h
+++ b/src/video_core/renderer_vulkan/vk_pipeline_cache.h
@@ -21,13 +21,11 @@
 #include "video_core/renderer_vulkan/fixed_pipeline_state.h"
 #include "video_core/renderer_vulkan/vk_graphics_pipeline.h"
 #include "video_core/renderer_vulkan/vk_renderpass_cache.h"
-#include "video_core/renderer_vulkan/vk_resource_manager.h"
 #include "video_core/renderer_vulkan/vk_shader_decompiler.h"
 #include "video_core/renderer_vulkan/wrapper.h"
 #include "video_core/shader/memory_util.h"
 #include "video_core/shader/registry.h"
 #include "video_core/shader/shader_ir.h"
-#include "video_core/surface.h"
 namespace Core {
 class System;
diff --git a/src/video_core/renderer_vulkan/vk_query_cache.cpp b/src/video_core/renderer_vulkan/vk_query_cache.cpp
index 813f7c162..bc91c48cc 100644
--- a/src/video_core/renderer_vulkan/vk_query_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_query_cache.cpp
@@ -4,7 +4,6 @@
 #include <algorithm>
 #include <cstddef>
-#include <cstdint>
 #include <utility>
 #include <vector>
diff --git a/src/video_core/renderer_vulkan/vk_query_cache.h b/src/video_core/renderer_vulkan/vk_query_cache.h
index b63784f4b..40119e6d3 100644
--- a/src/video_core/renderer_vulkan/vk_query_cache.h
+++ b/src/video_core/renderer_vulkan/vk_query_cache.h
@@ -5,7 +5,6 @@
 #pragma once
 #include <cstddef>
-#include <cstdint>
 #include <memory>
 #include <utility>
 #include <vector>
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.cpp b/src/video_core/renderer_vulkan/vk_rasterizer.cpp
index ccfd0e670..8b009fc22 100644
--- a/src/video_core/renderer_vulkan/vk_rasterizer.cpp
+++ b/src/video_core/renderer_vulkan/vk_rasterizer.cpp
@@ -9,14 +9,12 @@
 #include <vector>
 #include <boost/container/static_vector.hpp>
-#include <boost/functional/hash.hpp>
 #include "common/alignment.h"
 #include "common/assert.h"
 #include "common/logging/log.h"
 #include "common/microprofile.h"
 #include "core/core.h"
-#include "core/memory.h"
 #include "core/settings.h"
 #include "video_core/engines/kepler_compute.h"
 #include "video_core/engines/maxwell_3d.h"
@@ -301,7 +299,7 @@ RasterizerVulkan::RasterizerVulkan(Core::System& system, Core::Frontend::EmuWind
      buffer_cache(*this, system, device, memory_manager, scheduler, staging_pool),
      sampler_cache(device),
      fence_manager(system, *this, device, scheduler, texture_cache, buffer_cache, query_cache),
-      query_cache(system, *this, device, scheduler) {
+      query_cache(system, *this, device, scheduler), wfi_event{device.GetLogical().CreateEvent()} {
    scheduler.SetQueryCache(query_cache);
 }
@@ -575,6 +573,26 @@ void RasterizerVulkan::FlushAndInvalidateRegion(VAddr addr, u64 size) {
    InvalidateRegion(addr, size);
 }
+void RasterizerVulkan::WaitForIdle() {
+    // Everything but wait pixel operations. This intentionally includes FRAGMENT_SHADER_BIT because
+    // fragment shaders can still write storage buffers.
+    VkPipelineStageFlags flags =
+        VK_PIPELINE_STAGE_DRAW_INDIRECT_BIT | VK_PIPELINE_STAGE_VERTEX_INPUT_BIT |
+        VK_PIPELINE_STAGE_VERTEX_SHADER_BIT | VK_PIPELINE_STAGE_TESSELLATION_CONTROL_SHADER_BIT |
+        VK_PIPELINE_STAGE_TESSELLATION_EVALUATION_SHADER_BIT |
+        VK_PIPELINE_STAGE_GEOMETRY_SHADER_BIT | VK_PIPELINE_STAGE_FRAGMENT_SHADER_BIT |
+        VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT | VK_PIPELINE_STAGE_TRANSFER_BIT;
+    if (device.IsExtTransformFeedbackSupported()) {
+        flags |= VK_PIPELINE_STAGE_TRANSFORM_FEEDBACK_BIT_EXT;
+    }
+    scheduler.RequestOutsideRenderPassOperationContext();
+    scheduler.Record([event = *wfi_event, flags](vk::CommandBuffer cmdbuf) {
+        cmdbuf.SetEvent(event, flags);
+        cmdbuf.WaitEvents(event, flags, VK_PIPELINE_STAGE_TOP_OF_PIPE_BIT, {}, {}, {});
+    });
+}
 void RasterizerVulkan::FlushCommands() {
    if (draw_counter > 0) {
        draw_counter = 0;
@@ -895,6 +913,9 @@ void RasterizerVulkan::SetupVertexArrays(FixedPipelineState::VertexInput& vertex
 void RasterizerVulkan::SetupIndexBuffer(BufferBindings& buffer_bindings, DrawParameters& params,
                                        bool is_indexed) {
+    if (params.num_vertices == 0) {
+        return;
+    }
    const auto& regs = system.GPU().Maxwell3D().regs;
    switch (regs.draw.topology) {
    case Maxwell::PrimitiveTopology::Quads: {
diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.h b/src/video_core/renderer_vulkan/vk_rasterizer.h
index d41a7929e..0ed0e48c6 100644
--- a/src/video_core/renderer_vulkan/vk_rasterizer.h
+++ b/src/video_core/renderer_vulkan/vk_rasterizer.h
@@ -14,7 +14,6 @@
 #include <boost/functional/hash.hpp>
 #include "common/common_types.h"
-#include "video_core/memory_manager.h"
 #include "video_core/rasterizer_accelerated.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/renderer_vulkan/fixed_pipeline_state.h"
@@ -127,6 +126,7 @@ public:
    void SignalSyncPoint(u32 value) override;
    void ReleaseFences() override;
    void FlushAndInvalidateRegion(VAddr addr, u64 size) override;
+    void WaitForIdle() override;
    void FlushCommands() override;
    void TickFrame() override;
    bool AccelerateSurfaceCopy(const Tegra::Engines::Fermi2D::Regs::Surface& src,
@@ -276,6 +276,7 @@ private:
    vk::Buffer default_buffer;
    VKMemoryCommit default_buffer_commit;
+    vk::Event wfi_event;
    std::array<View, Maxwell::NumRenderTargets> color_attachments;
    View zeta_attachment;
diff --git a/src/video_core/renderer_vulkan/vk_sampler_cache.cpp b/src/video_core/renderer_vulkan/vk_sampler_cache.cpp
index 07bbcf520..e6f2fa553 100644
--- a/src/video_core/renderer_vulkan/vk_sampler_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_sampler_cache.cpp
@@ -2,11 +2,8 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <cstring>
-#include <optional>
 #include <unordered_map>
-#include "common/assert.h"
 #include "video_core/renderer_vulkan/maxwell_to_vk.h"
 #include "video_core/renderer_vulkan/vk_sampler_cache.h"
 #include "video_core/renderer_vulkan/wrapper.h"
@@ -42,9 +39,18 @@ VKSamplerCache::VKSamplerCache(const VKDevice& device) : device{device} {}
 VKSamplerCache::~VKSamplerCache() = default;
 vk::Sampler VKSamplerCache::CreateSampler(const Tegra::Texture::TSCEntry& tsc) const {
+    const bool arbitrary_borders = device.IsExtCustomBorderColorSupported();
+    const std::array color = tsc.GetBorderColor();
+    VkSamplerCustomBorderColorCreateInfoEXT border;
+    border.sType = VK_STRUCTURE_TYPE_SAMPLER_CUSTOM_BORDER_COLOR_CREATE_INFO_EXT;
+    border.pNext = nullptr;
+    border.format = VK_FORMAT_UNDEFINED;
+    std::memcpy(&border.customBorderColor, color.data(), sizeof(color));
    VkSamplerCreateInfo ci;
    ci.sType = VK_STRUCTURE_TYPE_SAMPLER_CREATE_INFO;
-    ci.pNext = nullptr;
+    ci.pNext = arbitrary_borders ? &border : nullptr;
    ci.flags = 0;
    ci.magFilter = MaxwellToVK::Sampler::Filter(tsc.mag_filter);
    ci.minFilter = MaxwellToVK::Sampler::Filter(tsc.min_filter);
@@ -59,7 +65,7 @@ vk::Sampler VKSamplerCache::CreateSampler(const Tegra::Texture::TSCEntry& tsc) c
    ci.compareOp = MaxwellToVK::Sampler::DepthCompareFunction(tsc.depth_compare_func);
    ci.minLod = tsc.GetMinLod();
    ci.maxLod = tsc.GetMaxLod();
-    ci.borderColor = ConvertBorderColor(tsc.GetBorderColor());
+    ci.borderColor = arbitrary_borders ? VK_BORDER_COLOR_INT_CUSTOM_EXT : ConvertBorderColor(color);
    ci.unnormalizedCoordinates = VK_FALSE;
    return device.GetLogical().CreateSampler(ci);
 }
diff --git a/src/video_core/renderer_vulkan/vk_scheduler.cpp b/src/video_core/renderer_vulkan/vk_scheduler.cpp
index ae7ba3eb5..82ec9180e 100644
--- a/src/video_core/renderer_vulkan/vk_scheduler.cpp
+++ b/src/video_core/renderer_vulkan/vk_scheduler.cpp
@@ -8,7 +8,6 @@
 #include <thread>
 #include <utility>
-#include "common/assert.h"
 #include "common/microprofile.h"
 #include "video_core/renderer_vulkan/vk_device.h"
 #include "video_core/renderer_vulkan/vk_query_cache.h"
diff --git a/src/video_core/renderer_vulkan/vk_scheduler.h b/src/video_core/renderer_vulkan/vk_scheduler.h
index 82a8adc69..970a65566 100644
--- a/src/video_core/renderer_vulkan/vk_scheduler.h
+++ b/src/video_core/renderer_vulkan/vk_scheduler.h
@@ -7,7 +7,6 @@
 #include <atomic>
 #include <condition_variable>
 #include <memory>
-#include <optional>
 #include <stack>
 #include <thread>
 #include <utility>
diff --git a/src/video_core/renderer_vulkan/vk_shader_decompiler.h b/src/video_core/renderer_vulkan/vk_shader_decompiler.h
index ffea4709e..f4c05ac3c 100644
--- a/src/video_core/renderer_vulkan/vk_shader_decompiler.h
+++ b/src/video_core/renderer_vulkan/vk_shader_decompiler.h
@@ -5,11 +5,7 @@
 #pragma once
 #include <array>
-#include <bitset>
-#include <memory>
 #include <set>
-#include <type_traits>
-#include <utility>
 #include <vector>
 #include "common/common_types.h"
diff --git a/src/video_core/renderer_vulkan/vk_shader_util.cpp b/src/video_core/renderer_vulkan/vk_shader_util.cpp
index 784839327..112df9c71 100644
--- a/src/video_core/renderer_vulkan/vk_shader_util.cpp
+++ b/src/video_core/renderer_vulkan/vk_shader_util.cpp
@@ -4,8 +4,7 @@
 #include <cstring>
 #include <memory>
-#include <vector>
-#include "common/alignment.h"
 #include "common/assert.h"
 #include "common/common_types.h"
 #include "video_core/renderer_vulkan/vk_device.h"
diff --git a/src/video_core/renderer_vulkan/vk_shader_util.h b/src/video_core/renderer_vulkan/vk_shader_util.h
index be38d6697..d1d3f3cae 100644
--- a/src/video_core/renderer_vulkan/vk_shader_util.h
+++ b/src/video_core/renderer_vulkan/vk_shader_util.h
@@ -4,7 +4,6 @@
 #pragma once
-#include <vector>
 #include "common/common_types.h"
 #include "video_core/renderer_vulkan/wrapper.h"
diff --git a/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h b/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h
index faf6418fd..3c4901437 100644
--- a/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h
+++ b/src/video_core/renderer_vulkan/vk_staging_buffer_pool.h
@@ -5,8 +5,6 @@
 #pragma once
 #include <climits>
-#include <unordered_map>
-#include <utility>
 #include <vector>
 #include "common/common_types.h"
diff --git a/src/video_core/renderer_vulkan/vk_texture_cache.cpp b/src/video_core/renderer_vulkan/vk_texture_cache.cpp
index de4c23120..55f43e61b 100644
--- a/src/video_core/renderer_vulkan/vk_texture_cache.cpp
+++ b/src/video_core/renderer_vulkan/vk_texture_cache.cpp
@@ -10,11 +10,9 @@
 #include <variant>
 #include <vector>
-#include "common/alignment.h"
 #include "common/assert.h"
 #include "common/common_types.h"
 #include "core/core.h"
-#include "core/memory.h"
 #include "video_core/engines/maxwell_3d.h"
 #include "video_core/morton.h"
 #include "video_core/renderer_vulkan/maxwell_to_vk.h"
@@ -26,7 +24,6 @@
 #include "video_core/renderer_vulkan/vk_texture_cache.h"
 #include "video_core/renderer_vulkan/wrapper.h"
 #include "video_core/surface.h"
-#include "video_core/textures/convert.h"
 namespace Vulkan {
diff --git a/src/video_core/renderer_vulkan/vk_texture_cache.h b/src/video_core/renderer_vulkan/vk_texture_cache.h
index 115595f28..f211ccb1e 100644
--- a/src/video_core/renderer_vulkan/vk_texture_cache.h
+++ b/src/video_core/renderer_vulkan/vk_texture_cache.h
@@ -7,19 +7,13 @@
 #include <memory>
 #include <unordered_map>
-#include "common/assert.h"
 #include "common/common_types.h"
-#include "common/logging/log.h"
-#include "common/math_util.h"
-#include "video_core/gpu.h"
-#include "video_core/rasterizer_cache.h"
 #include "video_core/renderer_vulkan/vk_image.h"
 #include "video_core/renderer_vulkan/vk_memory_manager.h"
 #include "video_core/renderer_vulkan/vk_scheduler.h"
 #include "video_core/renderer_vulkan/wrapper.h"
 #include "video_core/texture_cache/surface_base.h"
 #include "video_core/texture_cache/texture_cache.h"
-#include "video_core/textures/decoders.h"
 namespace Core {
 class System;
diff --git a/src/video_core/renderer_vulkan/vk_update_descriptor.h b/src/video_core/renderer_vulkan/vk_update_descriptor.h
index 6ba2c9997..cc7e3dff4 100644
--- a/src/video_core/renderer_vulkan/vk_update_descriptor.h
+++ b/src/video_core/renderer_vulkan/vk_update_descriptor.h
@@ -4,7 +4,6 @@
 #pragma once
-#include <type_traits>
 #include <variant>
 #include <boost/container/static_vector.hpp>
diff --git a/src/video_core/renderer_vulkan/wrapper.cpp b/src/video_core/renderer_vulkan/wrapper.cpp
index 7f5bc1404..2ce9b0626 100644
--- a/src/video_core/renderer_vulkan/wrapper.cpp
+++ b/src/video_core/renderer_vulkan/wrapper.cpp
@@ -87,6 +87,7 @@ void Load(VkDevice device, DeviceDispatch& dld) noexcept {
    X(vkCmdSetStencilReference);
    X(vkCmdSetStencilWriteMask);
    X(vkCmdSetViewport);
+    X(vkCmdWaitEvents);
    X(vkCreateBuffer);
    X(vkCreateBufferView);
    X(vkCreateCommandPool);
diff --git a/src/video_core/renderer_vulkan/wrapper.h b/src/video_core/renderer_vulkan/wrapper.h
index bda16a2cb..98937a77a 100644
--- a/src/video_core/renderer_vulkan/wrapper.h
+++ b/src/video_core/renderer_vulkan/wrapper.h
@@ -205,6 +205,7 @@ struct DeviceDispatch : public InstanceDispatch {
    PFN_vkCmdSetStencilReference vkCmdSetStencilReference;
    PFN_vkCmdSetStencilWriteMask vkCmdSetStencilWriteMask;
    PFN_vkCmdSetViewport vkCmdSetViewport;
+    PFN_vkCmdWaitEvents vkCmdWaitEvents;
    PFN_vkCreateBuffer vkCreateBuffer;
    PFN_vkCreateBufferView vkCreateBufferView;
    PFN_vkCreateCommandPool vkCreateCommandPool;
@@ -958,6 +959,15 @@ public:
        dld->vkCmdSetEvent(handle, event, stage_flags);
    }
+    void WaitEvents(Span<VkEvent> events, VkPipelineStageFlags src_stage_mask,
+                    VkPipelineStageFlags dst_stage_mask, Span<VkMemoryBarrier> memory_barriers,
+                    Span<VkBufferMemoryBarrier> buffer_barriers,
+                    Span<VkImageMemoryBarrier> image_barriers) const noexcept {
+        dld->vkCmdWaitEvents(handle, events.size(), events.data(), src_stage_mask, dst_stage_mask,
+                             memory_barriers.size(), memory_barriers.data(), buffer_barriers.size(),
+                             buffer_barriers.data(), image_barriers.size(), image_barriers.data());
+    }
    void BindTransformFeedbackBuffersEXT(u32 first, u32 count, const VkBuffer* buffers,
                                         const VkDeviceSize* offsets,
                                         const VkDeviceSize* sizes) const noexcept {
diff --git a/src/video_core/surface.cpp b/src/video_core/surface.cpp
index cc7181229..bbe93903c 100644
--- a/src/video_core/surface.cpp
+++ b/src/video_core/surface.cpp
@@ -145,6 +145,8 @@ PixelFormat PixelFormatFromRenderTargetFormat(Tegra::RenderTargetFormat format)
        return PixelFormat::RG8U;
    case Tegra::RenderTargetFormat::RG8_SNORM:
        return PixelFormat::RG8S;
+    case Tegra::RenderTargetFormat::RG8_UINT:
+        return PixelFormat::RG8UI;
    case Tegra::RenderTargetFormat::R16_FLOAT:
        return PixelFormat::R16F;
    case Tegra::RenderTargetFormat::R16_UNORM:
diff --git a/src/video_core/surface.h b/src/video_core/surface.h
index e0acd44d3..6da6a1b97 100644
--- a/src/video_core/surface.h
+++ b/src/video_core/surface.h
@@ -57,51 +57,52 @@ enum class PixelFormat {
    RGBA8_SRGB = 39,
    RG8U = 40,
    RG8S = 41,
-    RG32UI = 42,
+    RG8UI = 42,
-    RGBX16F = 43,
+    RG32UI = 43,
-    R32UI = 44,
+    RGBX16F = 44,
-    R32I = 45,
+    R32UI = 45,
-    ASTC_2D_8X8 = 46,
+    R32I = 46,
-    ASTC_2D_8X5 = 47,
+    ASTC_2D_8X8 = 47,
-    ASTC_2D_5X4 = 48,
+    ASTC_2D_8X5 = 48,
-    BGRA8_SRGB = 49,
+    ASTC_2D_5X4 = 49,
-    DXT1_SRGB = 50,
+    BGRA8_SRGB = 50,
-    DXT23_SRGB = 51,
+    DXT1_SRGB = 51,
-    DXT45_SRGB = 52,
+    DXT23_SRGB = 52,
-    BC7U_SRGB = 53,
+    DXT45_SRGB = 53,
-    R4G4B4A4U = 54,
+    BC7U_SRGB = 54,
-    ASTC_2D_4X4_SRGB = 55,
+    R4G4B4A4U = 55,
-    ASTC_2D_8X8_SRGB = 56,
+    ASTC_2D_4X4_SRGB = 56,
-    ASTC_2D_8X5_SRGB = 57,
+    ASTC_2D_8X8_SRGB = 57,
-    ASTC_2D_5X4_SRGB = 58,
+    ASTC_2D_8X5_SRGB = 58,
-    ASTC_2D_5X5 = 59,
+    ASTC_2D_5X4_SRGB = 59,
-    ASTC_2D_5X5_SRGB = 60,
+    ASTC_2D_5X5 = 60,
-    ASTC_2D_10X8 = 61,
+    ASTC_2D_5X5_SRGB = 61,
-    ASTC_2D_10X8_SRGB = 62,
+    ASTC_2D_10X8 = 62,
-    ASTC_2D_6X6 = 63,
+    ASTC_2D_10X8_SRGB = 63,
-    ASTC_2D_6X6_SRGB = 64,
+    ASTC_2D_6X6 = 64,
-    ASTC_2D_10X10 = 65,
+    ASTC_2D_6X6_SRGB = 65,
-    ASTC_2D_10X10_SRGB = 66,
+    ASTC_2D_10X10 = 66,
-    ASTC_2D_12X12 = 67,
+    ASTC_2D_10X10_SRGB = 67,
-    ASTC_2D_12X12_SRGB = 68,
+    ASTC_2D_12X12 = 68,
-    ASTC_2D_8X6 = 69,
+    ASTC_2D_12X12_SRGB = 69,
-    ASTC_2D_8X6_SRGB = 70,
+    ASTC_2D_8X6 = 70,
-    ASTC_2D_6X5 = 71,
+    ASTC_2D_8X6_SRGB = 71,
-    ASTC_2D_6X5_SRGB = 72,
+    ASTC_2D_6X5 = 72,
-    E5B9G9R9F = 73,
+    ASTC_2D_6X5_SRGB = 73,
+    E5B9G9R9F = 74,
    MaxColorFormat,
    // Depth formats
-    Z32F = 74,
+    Z32F = 75,
-    Z16 = 75,
+    Z16 = 76,
    MaxDepthFormat,
    // DepthStencil formats
-    Z24S8 = 76,
+    Z24S8 = 77,
-    S8Z24 = 77,
+    S8Z24 = 78,
-    Z32FS8 = 78,
+    Z32FS8 = 79,
    MaxDepthStencilFormat,
@@ -171,6 +172,7 @@ constexpr std::array<u32, MaxPixelFormat> compression_factor_shift_table = {{
    0, // RGBA8_SRGB
    0, // RG8U
    0, // RG8S
+    0, // RG8UI
    0, // RG32UI
    0, // RGBX16F
    0, // R32UI
@@ -269,6 +271,7 @@ constexpr std::array<u32, MaxPixelFormat> block_width_table = {{
    1,  // RGBA8_SRGB
    1,  // RG8U
    1,  // RG8S
+    1,  // RG8UI
    1,  // RG32UI
    1,  // RGBX16F
    1,  // R32UI
@@ -359,6 +362,7 @@ constexpr std::array<u32, MaxPixelFormat> block_height_table = {{
    1,  // RGBA8_SRGB
    1,  // RG8U
    1,  // RG8S
+    1,  // RG8UI
    1,  // RG32UI
    1,  // RGBX16F
    1,  // R32UI
@@ -449,6 +453,7 @@ constexpr std::array<u32, MaxPixelFormat> bpp_table = {{
    32,  // RGBA8_SRGB
    16,  // RG8U
    16,  // RG8S
+    16,  // RG8UI
    64,  // RG32UI
    64,  // RGBX16F
    32,  // R32UI
diff --git a/src/video_core/texture_cache/format_lookup_table.cpp b/src/video_core/texture_cache/format_lookup_table.cpp
index 25d2ee2e8..7032e0059 100644
--- a/src/video_core/texture_cache/format_lookup_table.cpp
+++ b/src/video_core/texture_cache/format_lookup_table.cpp
@@ -41,7 +41,7 @@ struct Table {
    ComponentType alpha_component;
    bool is_srgb;
 };
-constexpr std::array<Table, 76> DefinitionTable = {{
+constexpr std::array<Table, 77> DefinitionTable = {{
    {TextureFormat::A8R8G8B8, C, UNORM, UNORM, UNORM, UNORM, PixelFormat::ABGR8U},
    {TextureFormat::A8R8G8B8, C, SNORM, SNORM, SNORM, SNORM, PixelFormat::ABGR8S},
    {TextureFormat::A8R8G8B8, C, UINT, UINT, UINT, UINT, PixelFormat::ABGR8UI},
@@ -60,6 +60,7 @@ constexpr std::array<Table, 76> DefinitionTable = {{
    {TextureFormat::G8R8, C, UNORM, UNORM, UNORM, UNORM, PixelFormat::RG8U},
    {TextureFormat::G8R8, C, SNORM, SNORM, SNORM, SNORM, PixelFormat::RG8S},
+    {TextureFormat::G8R8, C, UINT, UINT, UINT, UINT, PixelFormat::RG8UI},
    {TextureFormat::R16_G16_B16_A16, C, SNORM, SNORM, SNORM, SNORM, PixelFormat::RGBA16S},
    {TextureFormat::R16_G16_B16_A16, C, UNORM, UNORM, UNORM, UNORM, PixelFormat::RGBA16U},
diff --git a/src/video_core/textures/decoders.cpp b/src/video_core/textures/decoders.cpp
index fae8638ec..548e4c3fe 100644
--- a/src/video_core/textures/decoders.cpp
+++ b/src/video_core/textures/decoders.cpp
@@ -382,4 +382,18 @@ std::size_t CalculateSize(bool tiled, u32 bytes_per_pixel, u32 width, u32 height
    }
 }
+u64 GetGOBOffset(u32 width, u32 height, u32 dst_x, u32 dst_y, u32 block_height,
+                 u32 bytes_per_pixel) {
+    auto div_ceil = [](const u32 x, const u32 y) { return ((x + y - 1) / y); };
+    const u32 gobs_in_block = 1 << block_height;
+    const u32 y_blocks = gob_size_y << block_height;
+    const u32 x_per_gob = gob_size_x / bytes_per_pixel;
+    const u32 x_blocks = div_ceil(width, x_per_gob);
+    const u32 block_size = gob_size * gobs_in_block;
+    const u32 stride = block_size * x_blocks;
+    const u32 base = (dst_y / y_blocks) * stride + (dst_x / x_per_gob) * block_size;
+    const u32 relative_y = dst_y % y_blocks;
+    return base + (relative_y / gob_size_y) * gob_size;
+}
 } // namespace Tegra::Texture
diff --git a/src/video_core/textures/decoders.h b/src/video_core/textures/decoders.h
index 9f2d6d308..06f3ebf87 100644
--- a/src/video_core/textures/decoders.h
+++ b/src/video_core/textures/decoders.h
@@ -59,4 +59,8 @@ void UnswizzleSubrect(u32 subrect_width, u32 subrect_height, u32 dest_pitch, u32
 void SwizzleKepler(u32 width, u32 height, u32 dst_x, u32 dst_y, u32 block_height,
                   std::size_t copy_size, const u8* source_data, u8* swizzle_data);
+/// Obtains the offset of the gob for positions 'dst_x' & 'dst_y'
+u64 GetGOBOffset(u32 width, u32 height, u32 dst_x, u32 dst_y, u32 block_height,
+                 u32 bytes_per_pixel);
 } // namespace Tegra::Texture
diff --git a/src/web_service/CMakeLists.txt b/src/web_service/CMakeLists.txt
index 0c9bb0d55..06ab7c59d 100644
--- a/src/web_service/CMakeLists.txt
+++ b/src/web_service/CMakeLists.txt
@@ -8,4 +8,4 @@ add_library(web_service STATIC
 )
 create_target_directory_groups(web_service)
-target_link_libraries(web_service PRIVATE common json-headers httplib lurlparser)
+target_link_libraries(web_service PRIVATE common nlohmann_json::nlohmann_json httplib lurlparser)
diff --git a/src/web_service/telemetry_json.cpp b/src/web_service/telemetry_json.cpp
index 7538389bf..7a480e33c 100644
--- a/src/web_service/telemetry_json.cpp
+++ b/src/web_service/telemetry_json.cpp
@@ -2,7 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <json.hpp>
+#include <nlohmann/json.hpp>
 #include "common/detached_tasks.h"
 #include "common/web_result.h"
 #include "web_service/telemetry_json.h"
diff --git a/src/web_service/verify_login.cpp b/src/web_service/verify_login.cpp
index ca4b43b93..bfaa5b70a 100644
--- a/src/web_service/verify_login.cpp
+++ b/src/web_service/verify_login.cpp
@@ -2,7 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <json.hpp>
+#include <nlohmann/json.hpp>
 #include "common/web_result.h"
 #include "web_service/verify_login.h"
 #include "web_service/web_backend.h"
diff --git a/src/yuzu/bootmanager.cpp b/src/yuzu/bootmanager.cpp
index 1cac2f942..3d759f77b 100644
--- a/src/yuzu/bootmanager.cpp
+++ b/src/yuzu/bootmanager.cpp
@@ -292,6 +292,8 @@ GRenderWindow::GRenderWindow(GMainWindow* parent_, EmuThread* emu_thread_)
    setLayout(layout);
    InputCommon::Init();
+    this->setMouseTracking(true);
    connect(this, &GRenderWindow::FirstFrameDisplayed, parent_, &GMainWindow::OnLoadComplete);
 }
@@ -385,6 +387,7 @@ void GRenderWindow::mousePressEvent(QMouseEvent* event) {
    } else if (event->button() == Qt::RightButton) {
        InputCommon::GetMotionEmu()->BeginTilt(pos.x(), pos.y());
    }
+    QWidget::mousePressEvent(event);
 }
 void GRenderWindow::mouseMoveEvent(QMouseEvent* event) {
@@ -397,6 +400,7 @@ void GRenderWindow::mouseMoveEvent(QMouseEvent* event) {
    const auto [x, y] = ScaleTouch(pos);
    this->TouchMoved(x, y);
    InputCommon::GetMotionEmu()->Tilt(pos.x(), pos.y());
+    QWidget::mouseMoveEvent(event);
 }
 void GRenderWindow::mouseReleaseEvent(QMouseEvent* event) {
diff --git a/src/yuzu/configuration/config.cpp b/src/yuzu/configuration/config.cpp
index a44eed047..75c6cf20b 100644
--- a/src/yuzu/configuration/config.cpp
+++ b/src/yuzu/configuration/config.cpp
@@ -743,6 +743,8 @@ void Config::ReadUIValues() {
    UISettings::values.profile_index = ReadSetting(QStringLiteral("profileIndex"), 0).toUInt();
    UISettings::values.pause_when_in_background =
        ReadSetting(QStringLiteral("pauseWhenInBackground"), false).toBool();
+    UISettings::values.hide_mouse =
+        ReadSetting(QStringLiteral("hideInactiveMouse"), false).toBool();
    ApplyDefaultProfileIfInputInvalid();
@@ -1169,6 +1171,7 @@ void Config::SaveUIValues() {
    WriteSetting(QStringLiteral("profileIndex"), UISettings::values.profile_index, 0);
    WriteSetting(QStringLiteral("pauseWhenInBackground"),
                 UISettings::values.pause_when_in_background, false);
+    WriteSetting(QStringLiteral("hideInactiveMouse"), UISettings::values.hide_mouse, false);
    qt_config->endGroup();
 }
diff --git a/src/yuzu/configuration/configure_general.cpp b/src/yuzu/configuration/configure_general.cpp
index 5ef927114..cb95423e0 100644
--- a/src/yuzu/configuration/configure_general.cpp
+++ b/src/yuzu/configuration/configure_general.cpp
@@ -26,6 +26,7 @@ void ConfigureGeneral::SetConfiguration() {
    ui->toggle_check_exit->setChecked(UISettings::values.confirm_before_closing);
    ui->toggle_user_on_boot->setChecked(UISettings::values.select_user_on_boot);
    ui->toggle_background_pause->setChecked(UISettings::values.pause_when_in_background);
+    ui->toggle_hide_mouse->setChecked(UISettings::values.hide_mouse);
    ui->toggle_frame_limit->setChecked(Settings::values.use_frame_limit);
    ui->frame_limit->setEnabled(ui->toggle_frame_limit->isChecked());
@@ -36,6 +37,7 @@ void ConfigureGeneral::ApplyConfiguration() {
    UISettings::values.confirm_before_closing = ui->toggle_check_exit->isChecked();
    UISettings::values.select_user_on_boot = ui->toggle_user_on_boot->isChecked();
    UISettings::values.pause_when_in_background = ui->toggle_background_pause->isChecked();
+    UISettings::values.hide_mouse = ui->toggle_hide_mouse->isChecked();
    Settings::values.use_frame_limit = ui->toggle_frame_limit->isChecked();
    Settings::values.frame_limit = ui->frame_limit->value();
diff --git a/src/yuzu/configuration/configure_general.ui b/src/yuzu/configuration/configure_general.ui
index 857119bb3..fc3b7e65a 100644
--- a/src/yuzu/configuration/configure_general.ui
+++ b/src/yuzu/configuration/configure_general.ui
@@ -72,6 +72,13 @@
            </property>
           </widget>
          </item>
+          <item>
+           <widget class="QCheckBox" name="toggle_hide_mouse">
+            <property name="text">
+             <string>Hide mouse on inactivity</string>
+            </property>
+           </widget>
+          </item>
         </layout>
        </item>
       </layout>
diff --git a/src/yuzu/main.cpp b/src/yuzu/main.cpp
index b44b4276c..86e8a1d49 100644
--- a/src/yuzu/main.cpp
+++ b/src/yuzu/main.cpp
@@ -135,6 +135,8 @@ __declspec(dllexport) int AmdPowerXpressRequestHighPerformance = 1;
 }
 #endif
+constexpr int default_mouse_timeout = 2500;
 constexpr u64 DLC_BASE_TITLE_ID_MASK = 0xFFFFFFFFFFFFE000;
 /**
@@ -236,6 +238,14 @@ GMainWindow::GMainWindow()
    // Show one-time "callout" messages to the user
    ShowTelemetryCallout();
+    // make sure menubar has the arrow cursor instead of inheriting from this
+    ui.menubar->setCursor(QCursor());
+    statusBar()->setCursor(QCursor());
+    mouse_hide_timer.setInterval(default_mouse_timeout);
+    connect(&mouse_hide_timer, &QTimer::timeout, this, &GMainWindow::HideMouseCursor);
+    connect(ui.menubar, &QMenuBar::hovered, this, &GMainWindow::ShowMouseCursor);
    QStringList args = QApplication::arguments();
    if (args.length() >= 2) {
        BootGame(args[1]);
@@ -1012,6 +1022,12 @@ void GMainWindow::BootGame(const QString& filename) {
    async_status_button->setDisabled(true);
    renderer_status_button->setDisabled(true);
+    if (UISettings::values.hide_mouse) {
+        mouse_hide_timer.start();
+        setMouseTracking(true);
+        ui.centralwidget->setMouseTracking(true);
+    }
    const u64 title_id = Core::System::GetInstance().CurrentProcess()->GetTitleID();
    std::string title_name;
@@ -1080,6 +1096,9 @@ void GMainWindow::ShutdownGame() {
        game_list->show();
    game_list->setFilterFocus();
+    setMouseTracking(false);
+    ui.centralwidget->setMouseTracking(false);
    UpdateWindowTitle();
    // Disable status bar updates
@@ -1837,6 +1856,15 @@ void GMainWindow::OnConfigure() {
    config->Save();
+    if (UISettings::values.hide_mouse && emulation_running) {
+        setMouseTracking(true);
+        ui.centralwidget->setMouseTracking(true);
+        mouse_hide_timer.start();
+    } else {
+        setMouseTracking(false);
+        ui.centralwidget->setMouseTracking(false);
+    }
    dock_status_button->setChecked(Settings::values.use_docked_mode);
    async_status_button->setChecked(Settings::values.use_asynchronous_gpu_emulation);
 #ifdef HAS_VULKAN
@@ -1970,6 +1998,30 @@ void GMainWindow::UpdateStatusBar() {
    emu_frametime_label->setVisible(true);
 }
+void GMainWindow::HideMouseCursor() {
+    if (emu_thread == nullptr || UISettings::values.hide_mouse == false) {
+        mouse_hide_timer.stop();
+        ShowMouseCursor();
+        return;
+    }
+    setCursor(QCursor(Qt::BlankCursor));
+}
+void GMainWindow::ShowMouseCursor() {
+    unsetCursor();
+    if (emu_thread != nullptr && UISettings::values.hide_mouse) {
+        mouse_hide_timer.start();
+    }
+}
+void GMainWindow::mouseMoveEvent(QMouseEvent* event) {
+    ShowMouseCursor();
+}
+void GMainWindow::mousePressEvent(QMouseEvent* event) {
+    ShowMouseCursor();
+}
 void GMainWindow::OnCoreError(Core::System::ResultStatus result, std::string details) {
    QMessageBox::StandardButton answer;
    QString status_message;
diff --git a/src/yuzu/main.h b/src/yuzu/main.h
index 0b750689d..60b17c54a 100644
--- a/src/yuzu/main.h
+++ b/src/yuzu/main.h
@@ -216,6 +216,8 @@ private:
    std::optional<u64> SelectRomFSDumpTarget(const FileSys::ContentProvider&, u64 program_id);
    void UpdateWindowTitle(const QString& title_name = {});
    void UpdateStatusBar();
+    void HideMouseCursor();
+    void ShowMouseCursor();
    Ui::MainWindow ui;
@@ -244,6 +246,7 @@ private:
    QString game_path;
    bool auto_paused = false;
+    QTimer mouse_hide_timer;
    // FS
    std::shared_ptr<FileSys::VfsFilesystem> vfs;
@@ -265,4 +268,6 @@ protected:
    void dropEvent(QDropEvent* event) override;
    void dragEnterEvent(QDragEnterEvent* event) override;
    void dragMoveEvent(QDragMoveEvent* event) override;
+    void mouseMoveEvent(QMouseEvent* event) override;
+    void mousePressEvent(QMouseEvent* event) override;
 };
diff --git a/src/yuzu/uisettings.h b/src/yuzu/uisettings.h
index a675ecf4d..830932d45 100644
--- a/src/yuzu/uisettings.h
+++ b/src/yuzu/uisettings.h
@@ -59,6 +59,7 @@ struct Values {
    bool confirm_before_closing;
    bool first_start;
    bool pause_when_in_background;
+    bool hide_mouse;
    bool select_user_on_boot;